teleconference on november 8, 2010. the appended...

This draft letter from the Ambient Air Monitoring and Methods Subcommittee will be discussed on a CASAC teleconference on November 8, 2010. The appended individual comments are from individual members of the

Ambient Air Monitoring and Methods Subcommittee and do not represent consensus CASAC advice or EPA policy. DO NOT CITE OR QUOTE. Updated on October 19, 2010

1 2 3 4 5 6 7 8 9

10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44

The Honorable Lisa P. Jackson Administrator U.S. Environmental Protection Agency 1200 Pennsylvania Avenue, N.W. Washington, D.C. 20460

Subject: Review of the “Near-road Guidance Document – Outline” and “Near-road Monitoring Pilot Study Objectives and Approach”

Dear Administrator Jackson: The Clear Air Scientific Advisory Committee (CASAC) Ambient Air Monitoring and Methods Subcommittee (AAMMS) met September 29-30, 2010 to review EPA’s Near-road Guidance Outline and Near-road Pilot Objectives. The impetus for the development of near-road monitoring guidance stems from the recent revisions to the primary National Ambient Air Quality Standards for NO2 in which EPA set out requirements for an NO2 monitoring network that includes monitors at locations where maximum NO2 concentrations are expected to occur, including within 50 meters of major roadways. While NO2 historically has been used as the indicator for ambient oxides of nitrogen (NOx), it is one of multiple gases (e.g. NO, NO2, HONO), CASAC notes that NO2 concentrations can significantly under represent NOx levels in the near-road environment. Although the requirement for near-road monitoring comes from the revised NO2 rule, EPA’s Office of Air Quality Planning and Standards (OAQPS) is developing guidance so that this expanded monitoring capability can be used for a variety of pollutants, pursuant to the Agency’s commitment to multi-pollutant monitoring approaches. Thus OAQPS sought CASAC’s advice in developing a guidance document for state and local monitoring agencies to implement the NO2 near-road network that is required to be operational by January 1, 2013. OAQPS also sought CASAC’s advice on the objectives and design of a near-road monitoring pilot study that is intended to allow EPA, state and local air monitoring agencies to evaluate and improve the near-road monitor siting process. CASAC believes EPA’s draft Guidance Outline provides the general factors that should be considered in developing a near road monitoring network, including those factors that should be considered in siting monitoring locations. A weakness in the current outline is that the objectives of the network are not well defined. The objectives and the associated reasoning should be clearly stated. With respect to EPA’s Near-road Monitoring Pilot Study Objectives, we note that the primary focus of the monitor site selection process is on annual average daily traffic

i



1 2 3 4 5 6 7 8 9

10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44

(AADT), however this data set may be overemphasized. Other factors to be considered include the physical characteristics (e.g. fleet mix, roadway design, congestion patterns, terrain, etc.), modeling information, and preliminary monitoring studies (e.g., multiscale/saturation studies). In fact, EPA should allow states and local agencies to bring as much information to bear on the site selection as is appropriate, with prioritization of installation of sites based on discussions between monitoring agencies and EPA offices. As a general matter, CASAC is deeply concerned about the timing proposed for the current network deployment, as well as for the Pilot Study. State and local resource constraints are another cause for concern. For all of the criteria pollutants as well as for the specific case of NO2, CASAC would like encourage EPA to commit the resources necessary to focus on the development and advancement of ambient air monitoring methods, with specific attention paid to assessment and possible modification or change of the Federal Reference and Equivalent Methods. We encourage evaluation of the use of a “true” NO2 monitor and inclusion of this instrument in the near-road monitoring program. We endorse the views expressed in National Association of Clean Air Agencies’ August 19, 2010, letter to Assistant Administrator Paul Anastas encouraging ORD “to increase its focus on the development and advancement of ambient air monitoring reference, equivalent and other sampling and analytical methods.” Measurement methods germane to near road monitoring mentioned in that letter include PM2.5, PM10-2.5, and ultrafine particles. We urge EPA to address biases in NO2 FRM measurements. We are concerned that the current time frame for the NO2 near-road network may not allow for adequate time to appropriately plan, execute and analyze the data from the Pilot Study to most effectively provide the information on which the near-road network can be designed. One way to mitigate the constraints of the current schedule would be to stage the deployment of the network, e.g., 10-20 sites the first year, 20-40 the next and the rest in the final year. Another approach would be to use near-road sites appropriately paired with sites from the National Core (NCore) monitoring network that is set to come online January 1, 2011 in the first round deployment. Consideration should be given to areas with NCore sites and larger populations. Such a staged approach would be consistent with the recommendations from the CASAC Oxides of Nitrogen Primary National Ambient Air Quality Standards (NAAQS) Review Panel (see Samet, Sept. 9, 2009), In this way, the network can evolve based on lessons learned from the Pilot Study as well as the operation of initial sites. CASAC is concerned that there could be a decrease in the number of population-oriented NO2 monitors in the new network. Many of the health studies that were used in the latest NAAQS NO2 review were based on the population oriented monitors of the current network. CASAC strongly recommends that a great majority of these monitors be maintained, particularly those that have been used in past health-focused studies, and that can be used to assess air quality trends.

ii



1 2 3 4 5 6 7 8 9

10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33

Just as we recommended a staged approach to the deployment of the near-road monitoring network, CASAC also recommends a tiered approach to the design of the near-road monitoring sites. A few sites should be comprehensively equipped such that they can provide much more detailed information about the composition of mobile source emissions and how that source changes with time and in response to controls. The bulk of the sites could be more modestly equipped. For example, the modestly equipped sites would [also] include black carbon, carbon monoxide (CO), and ultra-fine particulate matter (PM) monitoring capabilities responsive to the needs for assessing attainment with the applicable standards and the extent of near-road pollution exposure, as well as for use in health studies. We provide more detail in our response to the associated charge questions. CASAC wants to thank EPA for asking for input at this early stage in the process. The CASAC and AAMMS membership is listed in Enclosure A. CASAC’s consensus responses to the Agency’s charge questions are presented in Enclosure B. Individual review comments from the AAMMS are compiled in Enclosure C.

Sincerely, Dr. Armistead (Ted) Russell, Chair Dr. Jonathan M. Samet, Chair CASAC Ambient Air Monitoring & Clean Air Scientific Advisory Methods Committee Committee Enclosures

iii



1 2 3 4 5 6 7 8 9

10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44

U.S. Environmental Protection Agency

Clean Air Scientific Advisory Committee Ambient Air Monitoring and Methods Subcommittee for the

Review of Near-Road Monitoring to Support Measurement of Multiple National

Ambient Air Quality Standard (NAAQS) Pollutants CHAIR Dr. Armistead (Ted) Russell, Professor, Department of Civil and Environmental Engineering, Georgia Institute of Technology, Atlanta, GA MEMBERS Mr. George A. Allen, Senior Scientist, Northeast States for Coordinated Air Use Management (NESCAUM), Boston, MA Dr. Judith Chow, Research Professor, Desert Research Institute, Air Resources Laboratory, University of Nevada, Reno, NV Mr. Bart Croes, Chief, Research Division, California Air Resources Board, Sacramento, CA Dr. Kenneth Demerjian, Professor and Director, Atmospheric Sciences Research Center, State University of New York, Albany, NY Dr. Delbert Eatough, Professor of Chemistry, Department of Chemistry and Biochemistry , Brigham Young University, Provo, UT Dr. Eric Edgerton, President, Atmospheric Research & Analysis, Inc., Cary, NC Mr. Henry (Dirk) Felton, Research Scientist, Division of Air Resources, Bureau of Air Quality Surveillance, New York State Department of Environmental Conservation, Albany, NY Dr. Philip Hopke, Bayard D. Clarkson Distinguished Professor, Department of Chemical and Biomolecular Engineering, Clarkson University, Potsdam, NY Dr. Kazuhiko Ito, Assistant Professor, Department of Environmental Medicine, School of Medicine, New York University, Tuxedo, NY Dr. Peter H. McMurry, Professor, Department of Mechanical Engineering, University of Minnesota, Minneapolis, MN

iv



1 2 3 4 5 6 7 8 9

10 11 12 13 14 15 16 17 18

Mr. Richard L. Poirot, Environmental Analyst, Air Pollution Control Division, Department of Environmental Conservation, Vermont Agency of Natural Resources, Waterbury, VT Dr. Jay Turner, Associate Professor, Environmental & Chemical Engineering, Campus Box 1180 , Washington University , St Louis, MO Dr. Warren H. White, Research Professor, Crocker Nuclear Laboratory, University of California, Davis, CA Dr. Yousheng Zeng, Air Quality Services Director, Providence Engineering & Environmental Group LLC, Baton Rouge, LA SCIENCE ADVISORY BOARD STAFF Dr. Holly Stallworth, Designated Federal Officer, U.S. Environmental Protection Agency, Science Advisory Board, Washington, DC

v



1 2 3 4 5 6 7 8 9

10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36

U.S. Environmental Protection Agency Clean Air Scientific Advisory Committee

(CASAC)* CHAIR Dr. Jonathan M. Samet, Professor and Flora L. Thornton Chair, Department of Preventive Medicine, University of Southern California, Los Angeles, CA MEMBERS Dr. Joseph D. Brain, Cecil K. and Philip Drinker Professor of Environmental Physiology, Department of Environmental Health, Harvard School of Public Health, Harvard University, Boston, MA Dr. H. Christopher Frey, Professor, Department of Civil, Construction and Environmental Engineering, College of Engineering, North Carolina State University, Raleigh, NC Dr. Donna Kenski, Data Analysis Director, Lake Michigan Air Directors Consortium, Rosemont, IL Dr. Armistead (Ted) Russell, Professor, Department of Civil and Environmental Engineering, Georgia Institute of Technology, Atlanta, GA Dr. Helen Suh, Senior Lecturer on Environmental Chemistry and Exposure Assessment, Department of Environmental Health, School of Public Health, Harvard University, Boston, MA Dr. Kathleen Weathers, Senior Scientist, Cary Institute of Ecosystem Studies, Millbrook, NY SCIENCE ADVISORY BOARD STAFF Dr. Holly Stallworth, Designated Federal Officer, Science Advisory Board Staff Office, Washington, D.C.

* Roster to be updated.

vi



vii

1 2 3 4 5 6 7 8 9

10 11 12

NOTICE This report has been written as part of the activities of the EPA’s Clean Air Scientific Advisory Committee (CASAC), a federal advisory committee independently chartered to provide extramural scientific information and advice to the Administrator and other officials of the EPA. CASAC provides balanced, expert assessment of scientific matters related to issues and problems facing the Agency. This report has not been reviewed for approval by the Agency and, hence, the contents of this report do not necessarily represent the views and policies of the EPA, nor of other agencies within the Executive Branch of the federal government. In addition, any mention of trade names of commercial products does not constitute a recommendation for use. CASAC reports are posted on the EPA website at http://www.epa.gov/CASAC.



1 2 3 4

6 7 8 9

10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43

Consensus Response to Charge Questions on Near-Road Monitoring to Support Measurement of Multiple NAAQS Pollutants

(Charge Questions in Bold)

1. The accompanying draft guidance document outline provides an initial 5 thought of the major topics required in the near-road monitoring guidance that will aid state monitoring agencies in the identification and implementation of NO2 near road monitoring sites from a multi-pollutant perspective. Please comment on the overall content of the recommended topics in the draft outline. Please provide suggestions on any missing subjects that should be included in the guidance document and any unnecessary topics that are currently listed in the attached draft, if applicable.

It is not clear that EPA has identified the reasons (scientific objectives) for the multi-pollutant monitoring. We recommend that EPA clearly identify the monitoring and associated scientific objectives for the near-road monitoring program, and then build the document around these objectives. A possible set of objectives focused on NO2, for example, might be:

• Identification of hourly averaged NO2 by the most artifact free method available.

• Identification of hourly averaged concentrations of other NAAQS pollutants which accompany the observed NO2 concentrations.

• Identification of the atmospheric processes which contribute to the observed

concentrations of NO2.

• Identification of the sources which contribute to both measured concentrations of NO2 and the other measured NAAQS pollutants

The document should then discuss fully these objectives and the scientific basis for these objectives in the Introduction. This will, in turn, support the selection of the recommended pollutants to be monitored in the program and the protocols to be followed. In addition, we suggest the following more minor areas where the Guidance Document needs strengthening.

• General. The success of the near-road pilot project is critical to the future deployment of the 126 near-road monitoring sites by the end of 2012. Thus timing (and resources) are tight. EPA needs to focus on defining the objectives for the program and identify those key elements of the program essential to meeting those objectives. EPA needs to identify how to leverage existing sites,

viii



1 2 3 4 5 6 7

9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43

• Background. The background should document the scientific basis for the multi-8

pollutant objectives, as outlined in the bullets above. The literature review needs to be more comprehensive than at present. Some guidance on possible literature to site is included in individual responses. A conceptual model should be formed in this section that includes dispersion, deposition, chemical conversion and physical conversion. Special attention should be given to NO2 formation by the titration of O3 and NO2 depletion by photochemistry. The Background section should acknowledge that finding a site where NO2 and the various measured other pollutants all have the highest concentrations is not likely and compromises will be necessary and acceptable.

• Identifying Candidate Near-road Site Areas. We believe the criteria outlined for

this section are generally adequate for the identification of a site where maximum NO2 concentrations near a given near-road site may be determined. However, there is an apparent assumption that AADT is the primary siting criteria. The importance of meteorological and terrain variables should be emphasized. Further, since NO2 is a pollutant affected by both emissions from the roadway and nearby elevated pollutants (VOC, ozone, etc.), these factors should be considered in the site identification process. One critical element missing from the outline is the importance of the specific distance from the roadway chosen for study. The effect of differences in the gradient from the roadway of the multi pollutants studied needs to be acknowledged and discussed.

• Modeling. This section should include a discussion of the types of models

available and databases that can be used to inform siting process. The model should include both emissions and air quality models (dispersion).

• Monitoring. This section should provide a summary of methods used in past

roadside measurement studies, passive and active monitors that can be efficiently deployed at many locations and methods to interpret the data acquired. The emphasis at this level should be on the identification of high “true” NO2 hourly average concentrations. Care needs to be taken to insure that the use of mobile monitoring methods give results that do reflect the diurnal and seasonal locations of peak concentrations.

• Near-road Site Selection. The items outlined here seem reasonable. This section

ix



1 2

4 5 6 7

9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36

might draw from some of the existing guidance for sampler siting.

• Recommended Near-road Site Documentation. The adequacy of this section will 3 depend on the EPA objectives for the Pilot Study in addressing the appropriate multi-pollutant monitoring objectives. This will be better defined as the objectives of the near-road monitoring program are better defined.

2. EPA and NACAA envision the near-road guidance document to be written 8

from a multi-pollutant perspective. What pollutants and sub-species does the subcommittee believe should be included for consideration and discussion in the near-road monitoring guidance? Some potential species for consideration include NO2, NOX, NO, CO, PM (Ultrafine, 2.5, and 10), black carbon, air toxics (e.g., benzene, toluene, xylene, formaldehyde, acrolein, or 1, 3, butadiene), and ammonia. Please prioritize the recommended pollutants and provide the rationale for their ranking, including how this pollutant measurement will contribute to scientific and regulatory knowledge of near-road air quality and adverse human health effects.

The list and ranking of pollutants varied across the AAMMS members. Some questioned the usefulness of these indicators for the compliance purpose. The following ranking is based on the votes for the pollutants that had at least five votes from thirteen members of the AAMMS who submitted votes, taking into consideration the average ranking and its variability. The AAMMS believes meteorological parameters (wind speed and direction) should be one of the highest tier measurements considered as part of this network. Several members considered traffic counts as potentially important information. Other pollutants mentioned by some members include “true NO2” measured by techniques such as photolytic conversion difference, hourly metals by automated on-line XRF, and nephelometer measurement. AAMMS also recognizes that not all of the pollutants on the list can be measured at all the monitors because of the level of the cost and operational logistics. Therefore, the AAMMS recommends a tiered approach, in which a few of the sites would collect a more comprehensive set of pollutants, while a majority of them would measure only the first tier of pollutants in the list. We recommend that all of the data be obtained on an hourly averaged basis to allow intercomparison. Pollutant group Rationale and comments NO2, NO, NOx • Same instrument Black carbon • Diesel tracer

• Potential health relevance • Continuous methods available • Not a direct measure of carbon

CO • Gasoline vehicle tracer

x



• Dilution factors • Potential health relevance • Continuous methods available • Pending new NAAQS

Ultra-fine particles / particle number concentration

• Strong roadside gradient • Potential health relevance • Continuous methods available

Particle-size distribution • For a limited number of sites • Potential health relevance • More comprehensive health relevance than just particle number

PM10-2.5 • Potential health relevance • Re-suspended roadside particles • Speciation for metals • Important in future PM NAAQS reviews

PM2.5 • Continuous methods available • Potential health relevance • Speciation for metals, major components and organic marker compounds • Possibly FDMS to include semi-volatiles

EC/OC • Potential health relevance • Direct measure of carbon

CO2 • Normalization of pollutants to CO2 allows estimation of fuel-based emission factors • Could be important if signal large enough

Ozone • To understand photo-chemical processes • Not to be used for health effects analysis

NOy • Total nitrogen oxides and mass closure SO2 • To verify fuel content compliance

• Pending new NAAQS BTEX (benzene, toluene, ethylbenzene, and xylenes)

• Potential health relevance • Continuous methods available • Provides unique information as to how changing fuel characteristics and control systems are impacting emissions • Can provide additional info on which class of vehicle is impacting monitor

1 2

4 5 6

3. Identifying Candidate Near-road Site Areas 3 a. AADT & Fleet Mix – To consider fleet mix with regard to NO2, an

idea is to encourage states that have fleet mix information to take an approach that uses average, fleet-wide grams per mile emissions

xi



1 2 3 4 5 6 7 8 9

10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44

estimates (one for light duty vehicles and one for heavy duty vehicles), combined with AADT information to further weight which road segments in an area may be more conducive to produce peak pollutant concentrations. EPA would use the latest emission factor information to aid such a calculation. Given the variability in emission rates from on-road vehicles based on vehicle technology, fuel, speed, environmental conditions, etc., does the subcommittee believe this approach is an appropriate way to “consider” fleet mix in near-road site selection or is a more refined inventory and modeling analysis required?

b. AADT & Fleet Mix – Further, should the suggested approach above in question 4a to consider fleet mix via the use of average, fleet-wide emission factors, or the use of inventory and modeling analysis, take into account mobile source controls that are “on the books” but have not yet been fully realized due to fleet turnover? If so, how far out into the future should states consider their effects?

c. Roadway Design – Studies suggest and support the concept that roadway design influences pollutant dispersion near the road. The EPA suggests establishing sites at-grade with the road, without any nearby obstructions to air flow; however, the Agency recognizes that this might not always be feasible. Does the subcommittee agree with this recommendation for locating sites at-grade with no obstructions? What priority should be placed on this factor within the guidance, given the need for flexibility in identifying appropriate site locations?

d. Congestion Patterns – The congestion of a roadway can be estimated by the metric “Level of Service” (LOS). LOS uses a letter grade from A to F to identify a roadway’s performance, with “A” the best conditions where traffic flows at or above the posted speed limit and all motorists have complete mobility between lanes to “F” the worst congestion where travel time cannot be predicted and generally traffic demand exceeds the facility’s capacity. Since motor vehicles generally emit more pollutants during congestion operations (although noting that NOx and select other pollutant emissions can also increase with increasing speed), how important a parameter should LOS be in the determination of appropriate near-road monitoring sites? Does the subcommittee have a view on how reliable LOS estimates are across the country?

e. Terrain– State and local air agencies are required to consider terrain in the near-road monitoring site selection process, which in some cases may be inherently part of the roadway design. However, EPA recognizes that some states and local air agencies may have to make selections from amongst similar candidate sites that differ only by terrain, e.g. cut section versus open terrain, with or without vegetation, etc. Does the subcommittee agree that terrain and

xii



1 2 3 4 5 6 7 8 9

10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44

vegetation should be a consideration in the siting process? What priority should this parameter have in the overall process?

f. Meteorology – EPA took comment on, but did not finalize the requirement for near-road monitoring sites to be climatologically downwind of the target road segment. Reasons were because the additional limitations this would introduce in finding candidate sites would be in exchange for what may be a small increase in the opportunity to monitor peak NO2 concentrations. Further, with sites being within 50 meters of target road segments, the phenomenon of upwind meandering (pollutant transport upwind due to vehicle induced turbulence) further reduces that absolute need to be climatologically downwind. Finally, EPA recognized that, logically, the potential for peak NO2 concentration may very well occur when winds are calm or parallel (or nearly parallel) to the target road, allowing for pollutant build-up, as opposed to when winds are normal to the road. Although there is no requirement to be downwind, in the preamble to final NO2 NAAQS rule, EPA encouraged it when possible. EPA and NACAA intend to do the same in the guidance document. Does the subcommittee agree with this approach?

The specific factors identified in this charge question can all usefully inform the selection of near-road monitoring locations. An additional factor that may exhibit characteristic gradients in larger metropolitan areas is the middle-scale oxidant or odd-oxygen concentration, [O3]+[NO2]. A factor that needs emphasis is the role of freeways with high truck traffic volumes. As NOx and VOC emissions are continuously reduced over time, ambient oxidant levels will also go down, which means that titration of NO by ozone to form NO2 will become less important in comparison to direct NO2 emissions from new and retrofitted diesel trucks. NO2 emissions are highest during cruise mode, not congestion conditions for these trucks. CASAC recommends against an over-emphasis on AADT, noting for example that peak NO2 concentrations can occur in areas where widespread congestion limits AADT. To account for such interactions between different factors’ effects, the Agency should consider developing a conceptual or screening model to guide their integration. This would be intended as a tool to help rank candidate locations, and would not require quantitative concentration predictions. Recognizing the near-road NO2 measurement’s intended focus on NAAQS compliance rather than population exposure, CASAC stresses the importance of exposure in the overall balance of siting considerations. In response to specific elements of this charge question, CASAC’s recommendations are as follows.

a. States and cities vary considerably in the resolution of information available for fleet characteristics and fleet mix. States can be encouraged to use all available data in their planning, including local features such as truck and bus corridors.

xiii



2 3 4 5 6 7 8

10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43

b. Ambient monitoring is conducted in the ‘here and now’. Fleet turnover is too 1 slow to require consideration of its future effects in the initial planning. However, the transition to ‘cleaner’ diesels will have the effect of increasing primary tailpipe emissions of NO2.

c-e. CASAC agrees that roadway design, terrain, and congestion patterns all merit consideration. As noted above, it recommends that the interactions and tradeoffs among their effects be accounted for in a screening tool developed by the Agency for this purpose.

f. CASAC agrees that it is undesirable to site a near-road monitor 9 climatologically upwind.

4. Modeling is another tool that may be useful in the identification of candidate

near-road sites. In particular, the use of mobile source emissions modeling with MOVES and local-scale dispersion modeling with AERMOD, can be presented as part of the guidance document. Please comment on the available modeling tools, and their pros and cons, that the subcommittee believes may be appropriate to discuss and/or recommend for use in the near-road monitoring guidance document.

Like the various considerations discussed in Charge Question 3, modeling tools can also be used to inform the likely location of the maximum NO2 levels in an area. Taking a centralized approach, emissions and air quality models could be used to develop screening tools or screening criteria that incorporate many of the factors discussed in Charge Question 3. Alternatively, the models could directly be used by state and local agencies to inform the siting process. MOVES, along with other emissions inventory modeling tools, can provide valuable information about the spatial intensity of traffic-related NOx emissions, as well as NOx emissions from other sources. MOVES is a link-based model which has significant advantages for this application over the historical emissions models that were based on trip-average cycles. Air quality models, such as the CMAQ and CAMx chemical transport models and the AERMOD and CALINE dispersion models, can also be used to help identify candidate geographic areas within the metropolitan area and to compare and contrast candidate specific candidate monitoring sites, respectively. It is appreciated, however, that there significant uncertainties associated with the use of models for simulating the location of an extreme concentration statistic. The approaches used by AERMOD to simulate the conversion of NO to NO2 are relatively crude, and guidance should be given as to which approach should be used. State and local agencies should be encouraged to use spatially-resolved emissions estimates, and if readily available or available with relatively modest effort, air quality model results.

5. In regard to the process of identifying candidate near-road monitoring sites, beyond the evaluation of factors noted above in question 3, and the potential use of modeling, the use of saturation monitoring and on-road monitoring

xiv



1 2 3 4 5 6 7 8 9

10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44

are also possible tools that state and local air agencies may choose to utilize in the near-road site selection process.

“Multi-scale” monitoring is a more specific term than saturation monitoring. Before the pilot study or site selection, a literature review should be conducted that summarizes past roadside measurement and modeling studies, passive and active monitors that can be efficiently deployed at many locations, and methods to interpret the data acquired. Roadside monitoring for site selection and long-term monitoring would benefit from small, portable sensors that don’t require a large infrastructure (i.e., shelter, air conditioning, etc.) to operate. A pilot study plan should be assembled to clearly define the objectives of locating the sites (i.e., maximum hourly NO2 concentration), rationale for the selection of the measurements, measurement principles, concentration range, minimum detection limits, and potential interferences and biases. It should also elaborate on method intercomparison and modeling approaches. Sampling sites should be located at the prevailing downwind location in an array (e.g., to estimate perpendicular vs. effective distance from the road; Barzyk et al., 2009). The number of sampling sites would depend on the downwind distance in the micro- (10 – 100 m), middle- (100 – 500 m) and neighborhood- (500 m – 4 km) scales. Portable active or passive monitors that can be easily and inexpensively deployed would provide a good indication of where concentrations might be highest. Levels are likely to vary by season and often show an exponential decrease with distance from the curbside. Passive monitors using NO2-absorbing filters have some potential biases, but they have also been shown to be comparable and correlated with continuous measurements for integration times on the order of weeks. Passive samplers are a cost-effective and practical technology for mapping average spatial gradients as a prelude to sampler siting. However, the validity and sensitivity of the portable active or passive NO2 sampling systems for hourly measurements needs to be verified. To get better precision, collocated continuous monitors with duplicate passive samplers are needed at some of the sites. Past studies show an average coefficient of variation for duplicate passive NO2 measurements of 5 – 30%

a. If a state were inclined to use saturation monitoring to aid in the selection of a near-road monitoring site, and considering that the NO2 standard is a 1-hour daily maximum standard, what are the pros and cons to using passive devices to saturate an area to gather data?

Pros are low expense and operating cost. The major con is the much longer than one-hour averaging time, unless the passive device is modified to obtained active hourly data. The key is to avoid the need to relate long-duration passive sampler NO2 measurements to the one-hour NO2 NAAQS, which is defined as the maximum allowable concentration in an area (primarily near major roadways), expressed as a three-year average of the 98th percentile of the annual distribution of daily maximum one-hour concentrations. One

xv



1 2 3 4 5 6 7 8 9

10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44

option is to collect samples from sub-daytime periods (e.g., morning and evening rush hours) that are integrated over several days. A timer-based sampler may be used for this application. If many of the passive samplers are used for multi-scale monitoring it is important to position the sampling inlet at the same height (2 - 7 m above ground) at all sites

b. Likewise, what are the pros and cons to using non-passive devices, such

as near real-time or continuous devices including, but not limited to portable, non-FEM chemiluminescence methods for NO2 or Gas Sensitive Semiconductors (GSSs) for NO2 and other pollutants of interest?

Pros are short-duration samples, on the order of an hour or less. Cons are instrument procurement and operating expense, potentially higher than desired detection limits, and reliability of new technologies. There are several currently available or emerging technologies for microsensors. In addition to NO2, NO can often be obtained from these same sensors. Miniature sampling systems (e.g., 2B Technologies [Boulder, CO] NO monitor with NO to NO2 converter option and O3 monitor) have been used for remote monitoring, emission sampling, and unmanned aerial vehicles, and these might also be used for multi-scale monitoring These portable instruments need to be evaluated with regard to their sensitivity, stability, and accuracy. Data should be acquired over 1 min averages or less so that individual plumes can be detected.

c. Finally, what would be the pros and cons, to a state or local agency

attempting to use a specially outfitted vehicle to collect mobile measurements to assist in the near-road site selection process for NO2 specifically as well as other pollutants of interest?

Several mobile emissions systems or instrumented mobile vans have been applied to characterizing on-road and roadside concentrations. Pros are that these systems are moveable and obtain many different pollutant measurements. Unless state or local agencies are already equipped with a mobile sampling van, the disadvantages are the high cost of assembling or contracting these laboratories and the snapshot nature of their measurements. They usually need to be attended and can be parked for only a short time period. However, mobile monitoring systems are useful for site selection if they can be parked at the same location and perform sampling for a few days to one week. The state or local agencies may also consider using a fully-equipped trailer rather than a mobile van for siting. Eventually, the trailer can be converted to a permanent monitoring location.

6. EPA recognizes that CO concentrations are primarily influenced by gasoline vehicles as opposed to NO2 and PM2.5 concentrations, which are currently more heavily influenced by heavy-duty (diesel) vehicle emissions. If EPA were to propose a new set of minimum monitoring requirements for CO near roads, the near-road monitoring stations created under the

xvi



1 2 3 4 5 6 7 8 9

10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44

implementation of the NO2 monitoring requirements may be an advantageous infrastructure for state and local air agencies to leverage. However, EPA believes there are two issues not specifically considered in the near-road NO2 monitoring language that might influence where near-road CO monitors may be most appropriately placed. The two issues are 1) the consideration of where light duty vehicles are operating under ‘cold-start’ conditions, which may often not be on the larger arterials or highways in an area, and 2) the impacts of light duty vehicle congestion and idling in areas such as urban street canyons and/or urban cores.

a. Does the subcommittee believe that the light duty cold start and congestion factors will significantly influence the location of peak CO concentrations in an area? What priority should these factors be given when compared with the factors (AADT, Fleet Mix, Roadway Design, Congestion Patterns, Terrain, and Meteorology) already being considered for peak NO2?

The spatial distribution of cold start vehicles associated with urban commuting is in general broad and short term and their contribution to emissions associated with major highways adjacent to residential neighborhoods is likely small. To the extent cold start emissions contribute, they would mainly be in terms of CO and not NO2 concentrations and would not likely influence peak CO concentrations. That being said, at least one exception comes to mind. The departure of motor vehicles from major events (e.g. stadiums) where 20-30K vehicles may be simultaneously started and caught in congestion for 10s of minutes to an hour or more. The cold start contribution, here again, is limited in time but could contribute significantly as an emissions hot-spot impacting commuter exposes and concentrations in nearby neighborhoods. The prioritization of congestion and cold start factors relative to AADT, fleet mix, roadway design, terrain and meteorology should consider sensitivity analyses using line source models as outlined by the FHWA’s procedures for assessing traffic impacts for CO.

b. Does the subcommittee have an opinion on whether, and possibly how, these two issues of vehicles operating under cold start conditions and light duty vehicle congestion and idling in urban street canyons and/or urban cores be considered in a future, nationally applicable, CO monitoring proposal? Are there other factors that may affect peak CO concentrations and not affect peak NO2 concentrations that should also be considered for any future CO monitoring proposal?

Factors that affect near-road CO and NOx concentrations will be the distribution of gasoline and diesel vehicles. The primary NO2 fraction of NOx exhaust emissions will vary with engine type (gasoline spark ignition or diesel) and control equipment, e.g., three way catalyst and diesel particle filter trap technologies. CO emissions come principally from the gasoline engine exhaust and their peak concentrations will occur in the immediate vicinity of the emission source. While NO2 peak concentrations from

xvii



1 2 3 4 5 6 7 8 9

10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40

primary emissions will behave similarly to CO, NO2 peak concentrations from secondary reactions will lag behind as NO emissions react with entrained ozone to form NO2. The entrainment and transformation time afford the opportunity for the exhaust plume to transport and diffuse resulting in peak NO2 concentrations that are displaced in time and space as compared to CO. Idling in urban street canyons is problematic and should be discouraged. If, as in most cases, it is due to major traffic congestion it will result in high exposures involving commuters, pedestrians and local residence. Decisions to monitor at such locations must consider these exposures relative to other near-road exposure environments associated with high density population regions.

7. Does the committee believe that siting considerations for identifying the location of peak NO2 concentrations will likely address all of the high priority siting considerations for PM (particularly PM2.5) as well? If not, what other factors should be considered and what are the advantages in considering these factors for identifying the location of maximum PM concentration?

The majority of panelists acknowledged that there is an increment to PM2.5 in the near road environment. They did not suggest that the near road environment will represent the areas of highest PM2.5 concentration. This is due to the variety of sources of PM2.5 including the combination of primary emissions and secondary formation processes. The committee was also generally in agreement that mass based measurements and specifically the PM2.5 FRM is not appropriate for use at the near road NO2 sites. The PM2.5 FRM measurement has poor capture efficiency for the highly volatile emissions from mobile sources and is not suitable for the collection of hourly averaged data. The use of this method in the near road network would underestimate the significance of mobile sources and under predict the risk associated with this source of PM2.5. In general, the locations where maximum PM2.5 concentration are likely to be found include areas that are subject to regional transport and local stationary and area sources, are primarily urban, and are away from sinks of PM including ventilated roadways, heavy vegetation and water bodies. The Panel also discussed other fractions of PM in the near road environment including ultrafine particle count and size distribution and PMc. The siting for measurements of these size fractions are more suited to the near road environment. However, because the processes that affect downwind concentrations of NO2, UFP and PMc differ (Karner et al. 2010†), optimal downwind sites for those pollutants may also be different.

† Karner, A. A., Eisinger, D. S. and Niemeier, D. A. (2010). Near-Roadway Air Quality: Synthesizing the Findings from Real-World Data. Environ. Sci. Technol. 44:5334–5344.

xviii



1

3 4 5 6 7 8 9

10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33

8. In addition to PM2.5 mass, what other PM-related measurements are 2

desirable at near-road monitoring stations (e.g., UFP number, black carbon, EC/OC, PM coarse, etc.)?

Other PM measurements that should be considered include sub 100 nm ultrafine particles (UFP), BC, and speciated coarse particles. BC is an important vehicular emission that can be measured routinely in sampling networks. We recommend that it be measured. Measurements of UFP could be made either with a condensation particle counter (CPC) or an aerosol mobility spectrometer. A CPC measures the total concentration larger than the CPC’s minimum detectable size but provides no information about size. Mobility spectrometers provide information about particle concentration and size, which would be valuable for understating new particle formation and assessing health effects. Given current resource constraints, state and local agencies might be unable to collect and analyze data from mobility spectrometers. However, agencies might work collaboratively with interested universities. This would likely lead to a nested network design, with most stations using CPCs and a more limited number aerosol mobility spectrometers. Mobility distributions are currently being measured routinely in sampling networks worldwide. Vehicles emit coarse particles and resuspend coarse road dust that may be coated with toxic contaminants. We recommend measurements of the coarse particle metal content content on an hourly averaged basis, at least for the well instrumented sites. Questions regarding the monitor siting criteria for microscale CO, microscale PM2.5, and the new near-road NO2 siting criteria

9. To allow for near-road monitoring infrastructure to be multi-pollutant, and in reflection of the recently promulgated near-road NO2 siting criteria, reconsideration of the existing microscale CO siting criteria presented in sections 2, 6.2, and table E-4 in 40 CFR Part 58 Appendix E may be warranted. Does the subcommittee believe that reconsideration of microscale CO siting criteria is appropriate? Specifically, would an adjustment of CO siting criteria to match those of microscale PM2.5 and microscale near-road NO2 sites be logical and appropriate?

34 35 36 37 38 39 40 41 42 43 44

CASAC recommends that sampling criteria for CO and other monitors at sites installed to monitor near-road NO2 match those for NO2. The sampling configurations of existing micro-scale CO monitors should be assessed in terms of their own sampling objectives, and need not necessarily conform to those of near-road NO2 monitors.

xix

http://ecfr.gpoaccess.gov/cgi/t/text/text-idx?c=ecfr&sid=685b24c1eb28e918825b03d1f7b54a4f&rgn=div9&view=text&node=40:5.0.1.1.6.7.1.3.33&idno=40



10. Even if the adjustment of microscale CO siting criteria in sections 2, 6.2, and 1

table E-4 in 40 CFR Part 58 Appendix E to match that of microscale PM2.5 and microscale near-road NO2 is appropriate and proposed, should there be consideration to maintain the requirement on how urban street canyon or urban core microscale CO sites should be sited?

2 3 4 5 6 7 8 9

10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43

The Panel did not feel it had enough data to make a recommendation with respect to this charge question.

11. Does the subcommittee have an opinion on how “urban street canyons” or “urban core” might be defined, perhaps quantitatively, and with regard to use in potential rule language?

To define “urban street canyons” in relevant regulations, CASAC recommends that EPA consider the following factors:

• Traffic information such as AADT, fleet mix, posted speed limit or/and actual speed, traffic light cycle

• Street geometry o Ratio of the height of street side buildings to the width of the street (H/W

ratio). Some approaches need to be developed to treat the following scenarios for H/W ratio calculation

different heights of buildings on the two sides of the street – use the lower one or use the average. Using the lower one may be better.

tiered buildings – possibly use the average of all tiers, or use a graduated factor based on the angle between the vertical line and the line drawn along the tiered building – the larger the angle (more opening on the top of the street canyon), the lower the effective height (He) will be used in the H/W calculation.

o One-way vs. two-way street (more plug flow in a one-way street and more turbulent flow in a two-way street).

o Is the street lined with trees on the sidewalk? Tree canopy may have an effect of an umbrella and trap portion of pollutants at the street level.

o Slope of the street – higher vehicle emission on steeper streets unless it is a one-way street and the traffic direction is downhill.

o Some way to normalize the H/W ratio with respect to number of traffic lanes on the street. One approach would be modifying the H/W ratio to H*L/W, where L = number of lanes.

• Meteorological factors: the angle between the street and prevailing wind direction (higher concentrations are expected if the angle is 90 degree).

• Terrain: Frequency of calm conditions and/or drainage flow may influence concentrations.

xx




1 2 3 4 5 6 7 8 9

10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43

Although a set of cut-off values reflecting the above mentioned factors could be used to define urban street canyons, it is not advisable to set some clear-cut criteria (insufficient information exists to set clear-cut thresholds). It may be more appropriate to consider a street an urban street canyon if more than a certain number of these conditions are met. This will be a qualitative approach to define urban street canyons. For the definition of “urban core”, the CASAC has some suggestions for EPA to consider. EPA may use the U.S. Census Bureau definition of an “urban area” as a starting point. Per U.S. Census Bureau, an urban area is defined as "Core census block groups or blocks that have a population density of at least 1,000 people per square mile and surrounding census blocks that have an overall density of at least 500 people per square mile." Land features such as a river or a ridge may divide a CBSA into multiple urban cores. EPA may use population density to rank urban cores. EPA may further enhance this definition by factoring in “traffic density”, which could be calculated as the sum of AADT for every unit of road length (e.g., a length comparable to street block) in an area under consideration divided by the size of the area. Questions regarding the near-road monitoring pilot study

12. EPA and NACAA will select the locations for permanent sites that are part of the near-road pilot study based on which state or locals volunteer to participate and can process grant funds in a timely manner to deploy equipment. From this pool of volunteers, selection should be made on certain attributes that provide the best potential to fulfill pilot study objectives. In the attached draft white paper, EPA and NACAA have proposed some potential criteria for consideration in selecting where the fixed, permanent stations should be located. These considerations include choosing a large and a relatively small urban area based on population, an area with varied or complex terrain, an urban area with an operational NOX analyzer representative of neighborhood or larger spatial scales for comparison to the near-road NOX analyzer, and an urban area with a cooperative (or non-cooperative) Department of Transportation. Does the Subcommittee agree with these considerations? Further, are there other considerations that should be evaluated in selecting pilot cities to house permanent near-road monitoring stations as part of the pilot study?

Available funding constrains this NR pilot to only 2 or 3 sites unless EPA leverages existing sites and infrastructure that are NR-ish, possibly considering cities that are already conducting multi-pollutant assessment at multiple locations (Atlanta and New York City for example). Areas with existing "urban background" sites that have relevant NR pollutant measurements in place (that may include ozone for a measure of total urban oxidants) are desirable to assess the NR excess for key pollutants. The range of variables in this charge question cannot be fully evaluated with only a few sites. In reality, giving

xxi

http://en.wikipedia.org/wiki/Census_block



1 2 3 4 5 6 7 8 9

10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44

the severe constraints of funding and timing for the pilot, the siting decision may be driven largely by which S/L agencies have the resources to support the pilot work and where they can find and deploy a reasonable site quickly. With only 2-3 sites, it may be appropriate to choose "generic" sites (avoiding extremes of topography etc.) that are most likely to represent a large fraction of the final network and are near the middle of the 0-50 meter distance from the road (e.g., ~ 20-30 meters) and have HDD as a significant fraction of traffic. Reliance on AADT or an MSA's population are useful inputs but often may not be good indicators of maximum 1-hour NO2. The pilot fixed sites are not likely to inform how a wide range of siting characteristics would affect 1-hour NO2 concentrations. We might come closer to that goal by focusing the pilot on NO2 saturation studies with less emphasis (e.g. funding) on the fixed sites, but that is not a practical solution to the other goals of this pilot. There is some evidence suggesting that core urban zones not at large roadways may have the highest 1-hour NO2 values for some urban locations; saturation studies are ideal for assessing these sites. Cooperation of the local DOT may be useful for local traffic pattern characterizations. It may be worth encouraging academic or private sector groups to add in-kind supplemental measurements if that does not create multi-organization logistical issues. Finally, EPA, California Air Resources Board and several research groups have mobile monitors that could be deployed to quickly assess potential locations of highest 1-hour NO2 concentrations.

13. EPA and NACAA have proposed that at least two urban areas should have permanent near-road monitoring stations (that would fulfill NO2 near-road monitoring requirements) implemented for the pilot study. Please comment on the minimum equipment/pollutant measurement complement that should be deployed at each site and also the ideal equipment complement that each site should or could have, respectively. Specifically, what pollutants (e.g., NO2, NOX, NO, CO, PM (Ultrafine, 2.5, and 10), black carbon, air toxics (such as benzene, toluene, xylene, formaldehyde, acrolein, or 1,3, butadiene) and ammonia) and other information should the pilot study measure or gather at the fixed, permanent monitoring stations, and by what methods? This list should be in priority order, as feasible, and can include any NAAQS or non-NAAQS pollutant by any method (FRM/FEM and/or non-reference or equivalent methods), any particular type of other equipment for gathering supporting data such as meteorology or traffic counts.

The majority of AAMMS supports the deployment of at least two pilot study sites with a minimum measurement suite of NO2/NO/NOx , black carbon, particle number concentrations (UFP), CO, meteorological parameters, and traffic counts; and ideally the pollutant measurements tabulated in the response to Charge Question #2. Measurements for vehicle class and speed distributions are also encouraged, especially if a screening tool will be developed because this information can be used in its evaluation. If an FEM-approved photolytic NO2 monitor is available, its deployment along with an FRM NO2

xxii



1 2 3 4 5 6 7

9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44

monitor at each site is strongly encouraged. Other AAMMS members support the deployment of a single site with more extensive measurements to provide insights into key science questions about NO2. This would require a longer time commitment than currently programmed for the pilot study, but would ultimately better inform the regulatory process. Additional pilot study objectives could include the evaluation of passive sampling methods to be used in the saturation studies.

14. EPA and NACAA have proposed four to five urban areas to have saturation 8 monitoring, using either passive devices and/or continuous/semi-continuous saturation type multi-pollutant monitoring packages (i.e., several types of monitors in one mountable or deployable “package”). Please provide comment on:

a. The pollutants that should be measured with the saturation devices at

each saturation site. b. The number of saturation devices per pollutant, both passive and/or

continuous/semi-continuous, that may be deployed in each pilot city. c. Whether placing saturation monitoring devices near certain road

segments should include, at a minimum: 1) the highest AADT segment in an area, 2) the road segment with the highest number of heavy-duty truck/bus counts, 3) at a road segment with more unique roadway design, congestion pattern, or terrain in the area, and 4) if feasible, at a lower AADT segment with a similar fleet mix, roadway design, congestion, terrain, and meteorology as the top AADT road segment in the area.

The Subcommittee members expressed a wide range of views on the feasibility, ideal configuration(s) and potential usefulness of including a saturation monitoring component in the proposed pilot study. The pollutants sampled, number of sites per city, the number of cities and the kinds of sites for sampling need to be considered collectively, and in light of the limited budget and very short time frame available if the results are going to be of any use to the states in the establishment of permanent near-road sites. The kinds of sampling locations recommended by EPA and NACA are reasonable, but may be overly prescriptive, considering uncertainties in the kinds of sampling approaches that could actually be implemented. The simplest possible approach might be limited exclusively to the use of passive samplers and focused on NO2-only using Ogawa-type passive devices. Disadvantages of this approach include the single pollutant focus and longer-term cumulative nature of the resultant data. Assuming there are reasonable correlations between peak hourly concentrations and long term averages (as there have been at near-road sites in the UK), the sample aggregation of passive samplers may not be a major problem. Advantages of

xxiii



xxiv

1 2 3 4 5 6 7 8 9

10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

this simple approach include the very minimal siting constraints, the low (sampler, labor and analytical) cost per sample, which would allow deployment at a much larger array of locations, and the current availability of units with well characterized performance specifications. To support this simplified approach, it would be helpful if EPA would analyze data from existing NO2 sites that are located reasonably near major roads to determine how well weekly mean NO2 from multiple sites predicts the daily 1-hour maximum NO2 concentrations for the same time period. The more complex approach suggested by EPA would attempt to develop a portable compact “package” of active, continuous samplers for multiple pollutants of interest, including NO2, CO, PM2.5. Advantages include the ability to characterize and compare short-term hourly peak concentrations for different pollutants at different kinds of sites. Disadvantages include anticipated high cost per unit (fewer cities and sites), more constraining siting requirements (power, security), unproven track records for data quality, and anticipated time delays for equipment procurement, testing and field deployment. A third "intermediate" approach that CASAC recommends considering would be to combine passive samplers with timed, battery operated pumps which would draw a fixed flow rate of air through a small chamber housing containing passive NO2 samplers (and possibly other passive samplers for NOx, BC, O3) during specific time periods such as the morning rush hour(s) and afternoon hours of expected maximum secondary formation. The sample pump would improve the sensitivity and reduce the sample variability compared to purely passive devices, and would retain the advantages of relatively low cost and flexible siting locations of the passive samplers. However, this approach is much more complex than a simple passive sampler, and thus would require substantial resources to properly develop the method and evaluate its performance before it could be used. Without a significant amount of EPA technical support in method development and evaluation, the two more highly time-resolved saturation pilot study approaches described here are not practical; state and local air agencies do not have the resources to support this level of effort. We recommend that EPA provide appropriate resources to allow deployment of time-resolved saturation samplers such that these spatial studies can better inform the final network design. One possible alternative to use of saturation samplers at fixed locations would be to employ mobile sampling platforms to explore spatial and temporal patterns for multiple pollutants. While the high costs of such units precludes their development with available pilot study funds, it is possible that some states may already have such mobile units available, and the committee recommends that their use be considered as an alternative to fixed site sampling in locations where such units may be available.



1 2 3 4 5 6 7 8 9

10 11 12 13 14 15 16 17 18 19 20

Individual Comments from Ambient Air Monitoring and Methods Subcommittee

Mr. George Allen ................................................................................................................ 2 Dr. Judith Chow ................................................................................................................ 11 Mr. Bart Croes .................................................................................................................. 31 Dr. Kenneth Demerjian ..................................................................................................... 40 Dr. Delbert Eatough .......................................................................................................... 47 Mr. Dirk Felton ................................................................................................................. 55 Dr. Kazuhiko Ito ............................................................................................................... 63 Mr. Rich Poirot ................................................................................................................. 71 Dr. Jay Turner ................................................................................................................... 84 Dr. Warren H. White ......................................................................................................... 88 Dr. Yousheng Zeng ........................................................................................................... 94

1



1 2 3 4 5 6 7 8 9

10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44

Mr. George Allen There are two major topics covered by this Advisory: 1. How to best determine the likely location of the annual maximum value for urban-area 1-hour NO2 concentration to assist in the siting of the both the pilot and 126 site network monitors. 2. In addition to NO2, what other near-road (NR) relevant pollutants should be measured for both the pilot fixed-sites and the full network. EPA should be acknowledged for moving forward with a multi-pollutant NR network, an important initial step towards better understanding the exposure issues behind the observed NR health effects. Although the initial focus of this network is NO2, measurements of other non-NAAQS pollutants are critical for characterization of NR zones of influence. The Charge Questions are broken down into several categories, some of which are not well defined or have substantial content overlap: Guidance document development for the 126-site network CO and PM network issues related to NR monitoring Harmonization of siting criteria for NR multi-pollutant monitoring (including probe height) NR pilot studies -- saturation and fixed site designs: what to sample and where EPA assumes that these NO2 sites will all be near major roads, but existing data suggest that may not meet the network goals. Thus, the term near-road (NR) as used here also includes congested urban core areas that may be locations of maximum 1-hour NO2 concentrations. The NR pilot project is critical to informing the deployment of 126 NR NO2 monitoring sites by the end of 2012 (just over 2 years from now with siting plans due summer 2012). However, there is insufficient time to get and analyze all the data (saturation and fixed sites) from a pilot network; pilot studies must be done during both winter and summer seasons to account for potentially large seasonal NO2 variability; primary sources dominate in winter, with secondary sources a factor only during summer mid-day and afternoon when ozone may be present (not during morning rush-hour). EPA meeting materials (introduction to Charge Question 12) state that the saturation study should be performed before (to inform) deploying the fixed pilot sites, further extending the time needed to complete a proper pilot project. I am very concerned about the level of available funding for this pilot, which includes mulitple multi-pollutant sites, several saturation studies, urban background monitoring for pollutants of interest, and assumes substantial in-kind support from S/L agencies. A major challenge will be how to get useful information from this pilot effort with limited available resources ($800k) and a very short time-frame to adequately inform the larger NR network deployment. EPA needs to address who will do the data analysis for this

2



1 2 3 4 5 6 7 8 9

10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44

pilot (presumably a contractor), develop a plan for the analysis, and get external input on that plan. Because of these time constraints, if at all possible I recommend that the full network implementation be delayed or staggered or (ideally) both. If staggered, the first round of sites (10-15 in the largest urban areas?) could include additional (more intensive) measurements to better inform later phases. As EPA notes in their background material, there are many factors to consider for NR monitor siting; all of them can not adequately be addressed without a relatively large-scale pilot program. Thus, it may be useful to leverage existing sites that meet some of the NR siting requirements in the NO2 rule, and enhancing those existing sites, rather than deploying new sites. This would allow for more pilot sites and speed completion of the pilot study, but limit the “ideal” pilot siting design. I’d expect the saturation studies to be of more value re: informing the siting process, and the fixed sites to support the saturation studies and vet new methods. To support consideration of this approach, it would be very helpful if EPA could supply a list of existing NO2 sites that could at least loosely be considered NR (e.g., micro-scale siting), along with additional site meta-data such as AADT, vehicle type mix, other pollutants currently measured, and available matching urban background measurements. Finally, I strongly encourage EPA to evaluate the “NR” excess for key indicators (at a minimum: NO2, UFP and BC) as part of this pilot study. This requires a “matched set” of indicator measurements at the NR site and an appropriate “urban background” site - perhaps an NCore site if not too distant from the NR site. The “NR excess” metric allows an estimation of the gradient away from the road. Without the background measurements, the NR indicator data (BC, UFP) have no useful context. Charge Question 1. Content/Topics of Guidance Outline for full network siting and implementation. Since this is a brief outline, there is not a lot to comment on. 3b: AADT is over-emphasized as an initial step in site selection. It should be listed with the considerations in 3c. See other comments on AADT in response to charge question 3. 4. If modeling is going to be done by S/L agencies (that is my understanding), not EPA, it may have limited application. There is a wide range of expertise and resources across S/L air agencies for this kind of work. 5. Saturation monitoring may be the most effective approach for site identification, although the need for simplicity and low cost limits the use to multi-day samples rather than one to a few hours duration. 6. 6d seems to duplicate 3c.

3



1 2 3 4 5 6 7 8 9

10 11 12 13 14 15 16 17 18 19 20

Charge Question 2. “What pollutants and sub-species does the subcommittee believe should be included for consideration and discussion in the near-road monitoring guidance? Some potential species for consideration include NO2, NOX, NO, CO, PM (Ultrafine, 2.5, and 10), black carbon, air toxics (e.g., benzene, toluene, xylene, formaldehyde, acrolein, or 1, 3, butadiene), and ammonia. Please prioritize the recommended pollutants and provide the rationale for their ranking, including how this pollutant measurement will contribute to scientific and regulatory knowledge of near-road air quality and adverse human health effects.” For the full network, NO/NO2,UFP (CPC), BC, and CO, and meteorological measurements are the most important. NO2 and CO are NR-relevant NAAQS; no further discussion needed. UFP (particle # concentration by CPC) and BC are robust indicators of NR pollution influence and critical (along with NOx) to determine the “NR excess” pollution influence. PM2.5 should also be done, although it is generally not a good “indicator” of NR pollutants since the urban background is already elevated. The air toxics listed here are very useful but are a second-tier group. Ammonia is not of much interest despite the SCRT NOx controls now used for HDD. The list can be broken into tiers as follows:

21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37

Tier 1 - must do: For the 126 network site sites, NO/NO2, UFP (CPC), BC, CO, and meteorological measurements are the most important. NO2 and CO are NR-relevant NAAQS; no further discussion needed. UFP (particle # concentration by CPC) and BC are robust indicators of NR pollution influence and critical (along with NOx) to determine the "NR excess" pollution influence; # conc [UFP] has not been shown by itself to be of use for health effects yet. BC does show up as a significant predictor of a wide range of health outcomes, although it may [or may not] be acting as an indicator for other NR pollutants. Met (primarily wind) is critical to understanding temporal variations in NR pollutants. I would not recommend 3-d sonic for all sites [reliability issues at NR sites], despite the additional value those data have. Guidance for NO2 must include data acquisition of large negative values and address issues related to NO-NOx channel balance (matched gain) to avoid degradation of hourly NO2 data when NO is high and rapidly varying. Guidance for CPC/UFP must address diffusional losses of particles < ~25 nm in the sample inlet train.

38 Tier 2 - could do:

4



1 2 3 4 5 6 7 8 9

10

PM2.5 should also be done despite issues with the FRM/FEM methods; although it is generally not a good "indicator" of NR pollutants since the urban background is already elevated, it is a NAAQS that may often be highest at NR sites. The air toxics listed here are very useful for health impact, but are probably too expensive; acrolein and HCHO can not be done at hourly time-scales using practical methods. PM-coarse, and metals are doable at 1-h time-scales, but are resource intensive with health implications unclear. Paired indicator measurements at urban background site (BC, UFP) would be very useful for assessment of NR excess and estimation of the shape of the gradient away from the road, but maybe not everywhere.

11 12 13 14 15 16 17 18 19 20

Tier 3- could do at a small subset of sites [~10-15] to serve research needs: Hourly EC/OC, BTEX and other air toxics, SMPS or similar size-resolved UFP; hourly metals [automated on-line XRF]; true [photolytic] NO2, O3 [maybe NO chemiluminescent method to avoid measurement artifacts at low concentrations], CO2 [up and downwind], traffic counts or remote sensing of traffic. Paired indicator measurements at urban background site (BC, UFP). Dry nephelometers with pm-1.0 inlet cuts for fast-response PM-fine indicator. 3-D met at 3 and 10 meters. NOy paired with “true” photolytic NO2 may be useful for determination of total reactive nitrogen and NOz.

21 22 23 24 25 26 27 28 29 30 31

Don't do: Ammonia is not of much interest despite the SCRT NOx controls now used for HDD, and not practical at 1-hour resolution. Nitrate (no reason to expect it to be a NR issue, and little to no value for health effects); if NOy is done, ammonium nitrate is included in the NOz species. There was some discussion of commercial availability of a suitably robust UFP instrument (CPC) that could be successfully used in a large routine monitoring network. As of May 2010, a commercial water-based CPC is now available that appears to meet these needs; a brief evaluation of the shipping version (not pre-production) of this instrument is at:

32 33 34 35 36 37 38 39 40 41 42 43 44

http://home.comcast.net/~g_allen/TSI-3783-CPC_Allen.pdf Charge Question 3. Guidance for Identifying Candidate NR Sites for the full network AADT and fleet mix are two criteria of uncertain value. For NO2, HDD is the driving on-road primary source; highest NO2 might be found where there is a lot of HDD, significant congestion, and poor dispersion. If local ozone titration (summer only, secondary NO2) is a substantial driver of elevated NR NO2, the fleet mix becomes less important. AADT without related congestion data can be misleading; a free-flowing highway with high AADT is not likely to contribute to high NO2 levels. Terrain is a major factor; see work by Wang and Zhang, “Modeling Near-Road Air Quality Using a Computational Fluid Dynamics Model, CFD-VIT-RIT”, EST, 2009: 43

5

http://home.comcast.net/%7Eg_allen/TSI-3783-CPC_Allen.pdf



1 2 3 4 5 6 7 8 9

10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44

(7778–7783). Highest NR impact might be expected where the roadway is somewhat depressed relative to curbside terrain. Sound barriers and trees/vegetation (Baldauf-EPA work) may reduce NR impact for some parameters. Meteorology would ideally be measured at 10 meters to avoid the worst of road-induced turbulence (although at the time-scale of 1-hour, larger scale wind patterns would still likely dominate the wind data). It is not practical to require 3-d wind at all sites, although this could be useful. Sigma-Theta wind data may be useful as an indication of turbulence, as well as the different between 1-hour average scalar and resultant wind speed. Wind data should be sampled at 1-second intervals for these turbulence related metrics, and 5-minute averages may be very useful in a detailed assessment of periods of high concentration. For Up/Down wind siting, I recommend avoiding sites that are upwind relative to prevailing wind direction. The highest impact may occur where prevailing winds are parallel to the roadway, at which point there is no real up or down-wind side of the road. Charge Question 4. Use of modeling for NR Guidance Document This is not my area of expertise, although the concerns stated in Q1 part 4 remain, as well as the demonstrated ability of the models under consideration to reasonably predict locations of 1-hour maximum NO2 concentrations. Mobile source oriented models do not take micro-scale ozone titration into effect, and thus are useful only where primary NO2 would be expected to be the driver of 1-hour maximum NO2. Charge Question 5. Saturation and On-road Monitoring (Guidance Document) Saturation monitoring is a very useful tool to screen potential sites and learn more about the characteristics of sites with likely maximum NR impact. Ideally, a saturation study does not have to be constrained to NO2; with modest firmware modifications to existing personal monitors, BC can now be monitored for in these studies. While the BC (optical method) is highly time-resolved, the most practical method for NO2 remains the TEA-based passive samplers; these have been well characterized. They could be adapted for use in this work to measure only morning rush hour over a 5-weekday period by making the flow “active”: e.g., a timer, a suitable sampler “housing”, and a small (~ 200 ccm) pump. Some initial effort in modification and characterization of existing passive NO2 samplers would be needed. The passive samplers should be run at a fixed monitoring NR pilot site (with NO2 and BC) to validate field performance. The variability inherent in passive NO2 methods will be substantially reduced, since wind speed effects on “effective sampling rate” are essentially eliminated. These passive samplers typically need 50 ppb-hour of NO2 to provide stable data. With active sampling, this number drops, perhaps by 2 times. For a 15-hour (3h x 5days) sample period, once might expect useful data down to a few ppb. Ozone could also be added to any passive sampler-based approach. With NO and NO2, perhaps assisted by addition of BC and O3, the influence of primary vs. secondary NO2 can be assessed with morning rush-hour samples and

6



1 2 3 4 5 6 7 8 9

10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32

afternoon (~ 2-5pm local time) samples by looking at the ratios of these pollutants. This could answer one of the more complex questions for NO2 -- do primary or secondary process drive the maximum 1-h concentrations for a given site/season? For winter and morning rush-hour periods (minimal ozone), it is reasonable to assume that primary sources will dominate. For summer afternoons, that assumption can not be made. Taken together, this suggests that secondary processes are a potential driver of high NR NO2 only for summer mid-day and afternoon periods. Finally, in addition to official NWS data, reasonably local wind data must be collected for any saturation study. These data may be available from NOAA-MADIS meso-net sites; these sites have automated QC (done by MADIS) that can be used to assess data quality, and highly time-resolved (sub-hourly) historical data are available on the web. Other than for BC personal samplers using optical filter techniques, I do not recommend “on-line” methods for the saturation study. Cost, complicated siting logistics, and data quality are the primary concerns here. Having suggested this saturation study approach, the reality is that without significant EPA support, the best a S/L agency is likely to be able to do is a simple passive NO2 study, perhaps limited to weekday sampling rather than a full week. The Ogawa sampler can do NO2 and NOx in a single sampler, providing an indirect measurement of NO, which would be very useful. Since these samples would not be constrained to periods of likely NO2 maxima, existing NR-ish hourly data should be analyzed to determine how well a 5 or 7-day NO2 mean is correlated with the 1-h maximum for that period. This is an analysis EPA could easily do. I do not recommend on-road monitoring for this pilot project, in part because it is very resource intensive to do in a useful manner, and with the limited resources available a saturation study has more value with regard to the relevant siting questions especially when seasonality is considered. An exception could be if a local organization has a suitable mobile monitoring platform and the resources to deploy it. A recent example of mobile van spatial characterization of NR pollution is “Short-term variation in near-highway air pollutant gradients on a winter morning” (Durant et al., ACP, 2010; http://www.atmos-chem-phys.net /10/8341/2010/ 33

34 35 36 37 38 39 40 41 42 43 44

Charge Question 6. CO -- fleet mix, cold starts and urban canyons On-road sources of CO are different than NO2; there is essentially no CO from HDD. Thus, the areas of highest CO impact may be very different. Cold starts, idling, and fleet mix in urban canyons (and existing data) all suggest that a site for highest NO2 impact in an urban area may not be the location of highest CO impact. Finally, urban canyon siting is generally going to be very close to the curb (horizontally at least) -- typically < 5-10 meters. But this pilot can not afford to address the urban canyon CO issue. Charge Question 7. NR PM2.5 and PM10

7

http://home.comcast.net/%7Eg_allen/TSI-3783-CPC_Allen.pdf



1 2 3 4 5 6 7 8 9

10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44

Existing data has shown only a modest increment in NR PM2.5 in urban areas, primarily since the urban background is already elevated. PM2.5 is not generally a useful indicator of NR pollution excess gradients. Still, in urban areas without dominant industrial sources, the highest PM2.5 would normally be found near areas with substantial local traffic. PM10 would be expected to be somewhat higher at NR sites because of dust reintrainment; minimal hourly NR PM-coarse data exist, so it is difficult to assess this parameter at this time scale. Charge Question 8. Other PM-related measurements desirable at near-road monitoring stations See Q 2 above. It must be noted that for UFP (CPC particle # concentration), the inlet can not be at the height of NO2 and similar pollutants unless a carefully designed aerosol manifold is used. Diffusional losses of particles < ~20-30 nm can be large unless appropriate sample inlet trains are used, and it is common to see # concentration mode peaks in this size range at NR sites. Charge Question 9. Would an adjustment of CO siting criteria to match those of microscale PM2.5 and microscale near-road NO2 sites be logical and appropriate? Yes. Charge Question 10. Should there be consideration to maintain the requirement on how urban street canyon or urban core microscale CO sites should be sited? Yes. Charge Question 11. ...how “urban street canyons” or “urban core” might be defined, perhaps quantitatively, and with regard to use in potential rule language? These terms may be difficult to define for rule language across all urban areas. Right now, the approach seems to be “you know it when you see it.” Charge Question 12. ...potential criteria for consideration in selecting where the fixed, permanent [NR pilot] stations should be located. There will not be more than 2 or 3 of these sites in this NR pilot, as noted in the charge questions. Thus, the range of variables in this charge question can not be fully evaluated. In reality, giving the constraints of funding and timing noted above, the siting decision may be driven largely by what S/L agencies have the resources to support the pilot work and where they can find and deploy a reasonable site quickly. With only 2-3 sites, it may be best to choose “generic” sites that are most likely to represent a large fraction of the

8



1 2 3 4 5 6 7 8 9

10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44

final network. We are not going to learn how a wide range of siting characteristics would effect results from this pilot. We might come closer to that goal by constraining the pilot to saturation studies, with no fixed sites. But that is not a practical solution to the broader goals of this pilot. Cooperation of local the DOT may be useful, but the kind of data that needs to be collected for this pilot might be real-time remote traffic sensing that can estimate % large truck traffic -- data not usually available from a DOT. Charge Question 13. ...minimum equipment/pollutant measurement complement that should be deployed at each site and also the ideal equipment complement that each site should or could have... See Q 2 above, and my background comments about the importance of having matched urban-scale monitors for evaluation of the “NR excess” for key indicator measurements. It may also be useful to have real-time remote sensing of traffic count as used in the Las Vegas NR MSAT study -- this can also give information on large truck vs. other vehicle traffic. Charge Question 14. EPA is proposing that saturation studies for NO2 would be conducted in 4 or 5 urban areas. This is a reasonable goal. A) “The pollutants that should be measured with the saturation devices at each saturation site.” NO2 and maybe BC if resources allow. See Q 5 above. Met should be collected at one site in the area, possibly relying on existing MADIS meso-net data. B) “The number of saturation devices per pollutant, both passive and/or continuous / semi-continuous, that may be deployed in each pilot city.” I assume this Q is “how many sites?”. This is budget driven, but I suggest at least 6, with one of those being at a fixed site with robust NO2 and BC measurements (as a validation site). For those cities with a NR pilot fixed site, the collocation would be done there. For other cities, the most NR-ish site would be used for collocation of saturation study monitors. C) “Whether placing saturation monitoring devices near certain road segments should include, at a minimum: 1) the highest AADT segment in an area, 2) the road segment with the highest number of heavy-duty truck/bus counts, 3) at a road segment with more unique roadway design, congestion pattern, or terrain in the area, and 4) if feasible, at a lower AADT segment with a similar fleet mix, roadway design, congestion, terrain, and meteorology as the top AADT road segment in the area.” All except #4 above, as well as one or more urban canyon or similar non-highway site and a collocation site with regulatory monitors for QC use.

9



1

10



1 2 3 4 5 6 7

9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44

Dr. Judith Chow Subject: Preliminary Response to Charge Questions on NO2 Near Road Monitoring to Support Measurement of Multiple NAAQS Pollutants Date: September 20, 2010

1. Comment on the overall content of the recommended topics in the draft outline. 8 Provide suggestions on any missing subjects that should be included in the guidance document and any unnecessary topics that are currently listed in the attached draft, if applicable.

The Section 2 literature review needs to be more comprehensive than indicated. It should include a discussion of the relationships between the different pollutants at emission and the likely changes that they will experience with downwind transport from the roadway. It should emphasize the multipollutant (Chow et al., 2010a; Greenbaum and Shaikh, 2010; Hidy and Pennell, 2010; Mauderly et al., 2010) nature of near road exposure. A few reviews and meta-analyses of near-road concentrations have been published (Smichowski et al., 2008; Seigneur, 2009; Karner et al., 2010; Zhou and Levy, 2007) that can be used as starting points, although these are not specific to NO2 concentrations. A conceptual model should be formed in this section that includes dispersion, deposition, chemical conversion and physical conversion. Special attention should be given to NO2 formation by NO titration of O3 and NO2 depletion by photochemistry. Section 3 should contain a definition of source zones of influence and receptor zones of representation, defining middle-, neighborhood-, and urban scales (Chow et al., 2002). Compromises necessary to obtain multipollutant characterization should be defined. The list of variables seems complete, but there is an apparent assumption that AADT is the primary siting criteria. Meteorological and terrain variables also probably have important effects. One might find high levels in street canyons than on open roadways with good ventilation. Section 4 needs elaboration on the models to be considered and how their reported performance. There are several models that compare dispersion models with measurements, use new approaches such as computerized fluid dynamics (CFD) models to evaluate vehicle-induced turbulence and the effects of roadside obstructions, and that attempt to simulate chemical and physical transformations (Baik et al., 2007; Baker et al., 2004; Berkowicz et al., 2008; Buccolieri et al., 2009; Chakrabarty et al., 2000; Chan et al., 1995; Chang et al., 2009; Cheng et al., 2008; Cheng et al., 2009; Chu et al., 2005; Clarke et al., 2004; Di Sabatino et al., 2008; Dixon et al., 2006; Gidhagen et al., 2004b; Gidhagen et al., 2004a; Gokhale et al., 2005; Grawe et al., 2007; Gromke et al., 2008; Kang et al., 2008; Kondo et al., 2006; Kondo and Tomizuka, 2009; Kumar et al., 2009; Li et al., 2006; Liu and Leung, 2008; McNabola et al., 2009; Moussiopoulos et al., 2008; Murena et al., 2008; Murena et al., 2009; Ning et al., 2005; Oettl et al., 2006; Pohjola et al., 2003; Rodden et al., 1982; Sahlodin et al., 2007; Santiago and Martin, 2008; Solazzo et al., 2007; Tay et al., 2010; Tsai and Chen, 2004; Vardoulakis et al., 2002; Vardoulakis

11



1 2 3 4 5 6 7 8 9

10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

et al., 2003; Venkatram et al., 2007; Wang et al., 2006; Wang and Zhang, 2009; Xie et al., 2006; Yassin et al., 2008; Yassin et al., 2009; Yim et al., 2009; Zhou and Levy, 2008; Zhu and Hinds, 2005). In Section 5, “multi-scale” monitoring is a more specific term than saturation monitoring. This section should provide a summary of roadside measurement studies, passive and active monitors that can be efficiently deployed at many locations, and methods to interpret that data acquired. The conclusion might be that roadside monitoring for site selection and long-term monitoring needs small, portable sensors that don’t require a large infrastructure (i.e., shelter, air conditioning, etc.) to operate. Section 6 might draw from some of the existing guidance for sampler siting (U.S.EPA, 1997; U.S.EPA, 1998). Site documentation in Section 7 should include coordinates, photographs of the siting probe, and pictures of the surroundings.

2. What pollutants and sub-species should be included for consideration and discussion in the near-road monitoring guidance and what should be the priority of measurement?

See response to question 13.

3. What external variables should be used to identify candidate near-road monitoring sites?

a. Given the variability in emission rates from on-road vehicles based on vehicle technology, fuel, speed, environmental conditions, is the fleet mix in near-road site selection or is a more refined inventory and modeling analysis required?

Fleet mix is a good starting point, but this is likely to vary by time of day and the diurnal breakdown is unlikely to be available. Fleet mix should only be one variable considered in site selection.

b. Should the suggested approach consider fleet mix via the use of average, fleet-wide emission factors, or the use of inventory and modeling analysis, take into account mobile source controls that are “on the books” but have not yet been fully realized due to fleet turnover? If so, how far out into the future should states consider their effects?

Real-world emissions are likely to be quite different from certification-type emissions. It will take a long time for fleet evolution. Better to establish monitoring sites soon so that improvements can be tracked through long-term trends. It may be that the emission reduction measures that are “on the books” are not as effective as originally thought.

c. The EPA suggests establishing sites at-grade with the road, without any nearby obstructions to air flow; however, the Agency recognizes that this might not always be feasible. Does the subcommittee agree with this recommendation for locating sites at-grade with no obstructions? What priority should be placed on this factor within the guidance, given the need for flexibility in identifying appropriate site locations?

12



1 2 3 4 5 6 7 8 9

10 11 12 13 14 15 16 17 18 19 20 21 22

23 24 25 26 27

Higher concentrations will probably be found in more confined areas (e.g., street canyons) than near open roads with no obstructions. Obstructions between the vehicles and the monitors should be minimized.

d. How important a parameter should LOS be in the determination of appropriate near-road monitoring sites? Does the subcommittee have a view on how reliable LOS estimates are across the country?

Congestion varies throughout the day and on weekends vs. weekdays. LOS may provide a first-cut on roads that have congestion, but there is no evidence on how accurate it is as a congestion indicator.

e. Should terrain and vegetation should be a consideration in the siting process? What priority should this parameter have in the overall process?

Terrain is very important. Even small roadway dips can accumulate pollutants in hotspots (Bowen et al., 1993; Bowen and Egami, 1994).

f. Although there is no requirement to be downwind, in the preamble to final NO2 NAAQS rule, EPA encouraged it when possible. EPA and NACAA intend to do the same in the guidance document. Does the subcommittee agree with this approach?

No. Sampling should take place downwind of the prevailing wind. Even under stagnant conditions the ram effect of the vehicles will create flows parallel to the roadway. See Figure 6 pollution rose from Oettl et al. (2006), as reproduced below. Nothing is detected when the sampling location is upwind of the roadway.

(Oettl et al., 2006) Comment on the available modeling tools, and their pros and cons. See comments under Question 1. Roadway models must consider more than just linear dispersion of inert pollutants.

13



1 2 3 4 5 6 7 8 9

10 11 12 13 14 15 16 17

18 19 20 21 22 23 24 25 26 27 28

How might saturation and on-road monitoring be used for near-road site selection? Portable active or passive monitors that can be easily and inexpensively deployed would provide a good indication of where concentrations might be highest. Levels are likely to vary by season, as noted in Figure 1 of Zou et al. (2006), and show an exponential decrease with distance from the curbside. Passive monitors using NO2-absorbing filters have some potential biases, but have also been shown to be comparable with continuous measurements for integration times on the order of weeks (Ayers et al., 1998; Beckerman et al., 2008; Berkowicz et al., 2008; Crouse et al., 2009; De Fouquet et al., 2007; Douglas and Beaulieu, 1983; Faus-Kessler et al., 2008; Gilbert et al., 2003; Gonzales et al., 2005; Hauser et al., 2009; Heal and Cape, 1997; Heal et al., 1999; Heal et al., 2000; Henderson et al., 2007; Krochmal and Gorski, 1991; Mukerjee et al., 2004; Nash and Leith, 2010; Norris and Larson, 1999; Ozden and Dogeroglu, 2008; Parra et al., 2009; Piechocki-Minguy et al., 2006; Plaisance et al., 2004; Rava et al., 2007; Sekine et al., 2008; Shooter et al., 1997; Van Reeuwijk et al., 1998; Vardoulakis et al., 2009). This is probably a cost-effective and practical technology for mapping average spatial gradients as a prelude to sampler siting.

(Zou et al., 2006) Reprinted from Atmospheric Environment, 40, Zou et al., “Shifted power-law relationship between NO2 concentration and the distance from a highway: A new dispersion model based on the wind profile model, Copyright (2006) with permission from Elsevier. If a state were inclined to use saturation monitoring to aid in the selection of a near-road monitoring site, and considering that the NO2 standard is a 1-hour daily maximum standard, what are the pros and cons to using passive devices to saturate an area to gather data? Pros are low expense and operating cost. The major con is the long averaging time, much longer than 1 hour.

14



1 2 3 4 5 6 7 8 9

10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43

Likewise, what are the pros and cons to using non-passive devices, such as near real-time or continuous devices including, but not limited to portable, non-FEM chemiluminescence methods for NO2 or Gas Sensitive Semiconductors (GSSs) for NO2 and other pollutants of interest? Pros are short-duration samples, on the order of an hour or less. Cons are instrument procurement and operating expense, potentially higher than desired detection limits, and reliability of new technologies. Finally, what would be the pros and cons, to a state or local agency attempting to use a specially outfitted vehicle to collect mobile measurements to assist in the near-road site selection process for NO2 specifically as well as other pollutants of interest? Several mobile emissions systems have been applied to characterizing on-road and roadside concentrations (Bukowiechi et al., 2002; Bukowiecki et al., 2003; Cocker et al., 2004a; Cocker et al., 2004b; Durbin et al., 2007; Herndon et al., 2005; Isakov et al., 2007; Kittelson et al., 2004; Kittelson et al., 2006; Morawska et al., 2007; Nussbaum et al., 2009; Pirjola et al., 2004). Pros are that these systems are moveable and obtain many different pollutant measurements. Cons are the large cost of assembling or contracting these laboratories and the snapshot nature of their measurements, as they usually need to be attended and can be parked for only a short time period. To what extent will light duty cold start and congestion factors will significantly influence the location of peak CO concentrations in an area? The cold start segment has been found to affect emissions for many pollutants, including CO (Cadle et al., 2001; Chan and Zhu, 1999; Chase et al., 2000; Cook et al., 2007; Cotte et al., 2001; Gullett et al., 2006; Huai et al., 2004; Joumard and Andre, 1990; Joumard et al., 2000; Kittelson et al., 2006; Korin et al., 1999; Lenaers, 1996; Lough et al., 2005; Ludykar et al., 1999; Maricq et al., 1999; Mathis et al., 2005; Pornet et al., 1995; Ristimaki et al., 2005; Schauer et al., 2008; Singer et al., 1999; Weilenmann et al., 2005; Weilenmann et al., 2009; Westerholm et al., 1996). This is of fairly short duration (minutes) and would most likely affect emissions in garages, driveways, parking lots and side streets rather on the heavily-travelled thoroughfares. If it is desired to characterize cold starts, sampling systems should be located near where cars turn onto major arteries from nearby neighborhoods. Even so, only those living most closely to the intersection will exhibit cold start emissions. What priority should these factors be given when compared with the factors (AADT, Fleet Mix, Roadway Design, Congestion Patterns, Terrain, and Meteorology) already being considered for peak NO2? Priority should be low. Do these two issues of vehicles operating under cold start conditions and light duty vehicle congestion and idling in urban street canyons and/or urban cores be considered in a future, nationally applicable, CO monitoring proposal? This is probably better treated as an emission standard that would minimize cold start emissions through technological means. Are there other factors that may affect peak CO concentrations and not affect peak NO2 concentrations that should also be considered for any future CO monitoring proposal?

15



1 2 3 4 5 6 7 8 9

10 11 12 13 14 15 16 17 18 19 20

CO is relatively inert and is often used to normalize other pollutants for dispersion downwind of a roadway (Zhang and Wexler, 2004). It is expected that NO2 emissions will disperse in a similar manner, although they still experience transformation processes that differ from those of CO. Will siting considerations for identifying the location of peak NO2 concentrations address all of the high priority siting considerations for PM (particularly PM2.5) as well? If not, what other factors should be considered and what are the advantages in considering these factors for identifying the location of maximum PM concentration? No. PM2.5 is a combination of primary and secondary particles from a wide variety of emission sources. Roadside sampling is useful for characterizing the motor vehicle contribution, but it may bias the urban- and regional-scale PM2.5 compositions and exposures. These monitors would be considered Special Purpose Monitors according to the PM siting criteria (U.S.EPA, 1997). In addition to PM2.5 mass, what other PM-related measurements are desirable at near-road monitoring stations (e.g., UFP number, black carbon, EC/OC, PM coarse, etc.)? Particle size distribution and particle number by continuous methods. PM speciation on filters, including elements, ions, OC/EC, and organic markers, would be useful to develop source profiles for emission inventory speciation and receptor modeling. To allow for near-road monitoring infrastructure to be multi-pollutant, and in reflection of the recently promulgated near-road NO2 siting criteria, reconsideration of the existing microscale CO siting criteria presented in sections 2, 6.2, and table E-4 in 40 CFR Part 58 21

22 23 24 25 26 27 28 29 30 31 32 33

Appendix E may be warranted. Does the subcommittee believe that reconsideration of microscale CO siting criteria is appropriate? Specifically, would an adjustment of CO siting criteria to match those of microscale PM2.5 and microscale near-road NO2 sites be logical and appropriate? 40CFRPart58 Appendix E calls for CO roadside monitors to be located at 2 to 10 m from the nearest traffic lane for open roads. In a street canyon, the monitor is to be at least 10 m from an intersection. An NO2 monitor might register higher concentrations near the 10 m downwind location owing to NO2 formation by reaction of the NO2 with O3. It seems that a reasonable compromise on the setback could be derived that would serve both purposes. A more detailed examination of NO2, NO, CO, and O3 data is needed in the pilot study to better determine the optimum distance from the roadside. Even if the adjustment of microscale CO siting criteria in sections 2, 6.2, and table E-4 in 40 CFR Part 58 Appendix E to match that of microscale PM2.5 and microscale near-road NO2 is appropriate and proposed, should there be consideration to maintain the requirement on how urban street canyon or urban core microscale CO sites should be sited?

34 35 36 37 38 39 40 41 42 43

The NO2 and CO siting criteria should be the same. There is more to be gained from the multipollutant measurements than is lost by slight differences in maximum hourly concentrations. Does the subcommittee have an opinion on how “urban street canyons” or “urban core” might be defined, perhaps quantitatively, and with regard to use in potential rule language?

16






1 2 3 4 5 6 7 8 9

10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43

There is a reasonable literature on measurements in street canyons (Baik and Kim, 2002; Bakeas and Siskos, 2003; Boddy et al., 2005a; Boddy et al., 2005b; Buccolieri et al., 2010; Cai et al., 2008; Caton et al., 2003; Chan et al., 2003; Cheng et al., 2009; Chu et al., 2005; Di Sabatino et al., 2008; Dobre et al., 2005; Eliasson et al., 2006; Genikhovich et al., 2005; Gromke et al., 2008; Hang et al., 2009; Kassomenos et al., 2004; Kim and Baik, 2004; Kumar et al., 2008; Kumar et al., 2009; Lam et al., 2008; Li et al., 2005; Li et al., 2009; Longley et al., 2003; Longley, 2004; Longley et al., 2004; McNabola et al., 2009; Molina, 1996; Moussiopoulos et al., 2008; Murena and Vorraro, 2003; Murena et al., 2008; Prajapati et al., 2009; Santiago and Martin, 2008; Scaperdas and Colvile, 1999; So et al., 2005; Stein and Toselli, 1996; Tay et al., 2010; Tsai et al., 2005; Venegas and Mazzeo, 2000; Voigtlander et al., 2006; Xie et al., 2003; Xie et al., 2005; Xie et al., 2006; Xie et al., 2007; Yassin et al., 2009), supplementing the street canyon modeling literature cited in the response to Question 1. These measurement and modeling studies need to be critically evaluated to answer this question. A quick survey suggests that there are various degrees of roadside obstructions that will have large effects on concentrations. To what extent are the pilot study site selection criteria of a large and a relatively small urban area based on population, an area with varied or complex terrain, an urban area with an operational NOX analyzer representative of neighborhood or larger spatial scales for comparison to the near-road NOX analyzer, and an urban area with a cooperative (or non-cooperative) Department of Transportation complete and adequate?. Additional criteria should include periods of morning stagnation and low inversion, differing morning O3 levels that might enhance NO2 through NO titration, cold as well as warm environments that might experience different emission levels owing to cold starts. Comment on the minimum equipment/pollutant measurement complement that should be deployed at each site and also the ideal equipment complement that each site should or could have, respectively. Specifically, what pollutants (e.g., NO2, NOX, NO, CO, PM (Ultrafine, 2.5, and 10), black carbon, air toxics (such as benzene, toluene, xylene, formaldehyde, acrolein, or 1,3, butadiene) and ammonia) and other information should the pilot study measure or gather at the fixed, permanent monitoring stations, and by what methods? This list should be in priority order, as feasible, and can include any NAAQS or non-NAAQS pollutant by any method (FRM/FEM and/or non-reference or equivalent methods), any particular type of other equipment for gathering supporting data such as meteorology or traffic counts. As implied in the charge questions, a specialized multi-pollutant monitoring package should be assembled and applied in these studies. There are several examples of such packages that have been assembled for neighborhood-scale studies, on-board emissions sampling, and unmanned aerial vehicles that have contain potentially applicable sensors, but these would need to be evaluated with respect to their sensitivity, stability, and accuracy. Data should be acquired over 1 min averages or less so that individual plumes can be detected. Nitrogen dioxide (NO2). This is top priority because it is the focus of the study. There are several currently available or emerging technologies for microsensors (Brunet et al., 2008; Currie et al., 1999; Egashira et al., 1996; Forleo et al., 2005; Gurlo et al., 1998; Oto

17



1 2 3 4 5 6 7 8 9

10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44

et al., 2001; Sitnikov et al., 2005; Talazac et al., 2001). NO can often be obtained from these same sensors. Carbon dioxide (CO2): Normalizing other pollutants to CO2 allows fuel-based emission factors to be developed (Kean et al., 2000; Sawyer et al., 2000). Commercially available microchip IR sensors are available for CO2 measurements (Chow et al., 2010b). Black Carbon (BC): On a short-duration minute basis, this would allow cold starts and diesel exhaust to be separated from others and related to the NO2 emissions from individual vehicle plumes. A portable aethalometer (Hansen and Mocnik, 2010) is available for filter transmission measurements of BC, and more portable photoacoustic measurement systems (Kok and Baumgardner, 2010) are emerging. Carbon Monoxide (CO): CO is a priority pollutant and is an indicator of gasoline engine contributions, especially for cold starts and poorly maintained engines. Several small detectors are available (Do and Chen, 2007; Oto et al., 2001). Ozone (O3): This would be important for estimating NO titration to NO2. Several microsensors are available or are emerging technologies (Do and Chen, 2007; Gurlo et al., 1998; Ulanovsky et al., 2001; Vallejos et al., 2007). Particle number: This would indicate a potential adverse health effect. Portable CPC counters are available. PM10 and PM2.5: Coarse particles (PM10-2;5) may be affected by road dust while PM2.5 is largely from vehicle exhaust. Optical particle counters can provide a real-time surrogate for these components (Wang et al., 2009; Cheng, 2008; Heim et al., 2008; Linnainmaa et al., 2008). EPA and NACAA have proposed four to five urban areas to have saturation monitoring, using either passive devices and/or continuous/semi-continuous saturation type multi-pollutant monitoring packages (i.e., several types of monitors in one mountable or deployable “package”). Please provide comment on: The pollutants that should be measured with the saturation devices at each saturation site. See answers to Question 13. The number of saturation devices per pollutant, both passive and/or continuous/semi-continuous, that may be deployed in each pilot city. Four sampling systems should be located downwind of the roadway at various distances. One should be located in a neighborhood near the road and one should be located upwind of the urban area. Whether placing saturation monitoring devices near certain road segments should include, at a minimum: 1) the highest AADT segment in an area, 2) the road segment with the highest number of heavy-duty truck/bus counts, 3) at a road segment with more unique roadway design, congestion pattern, or terrain in the area, and 4) if feasible, at a lower AADT segment with a similar fleet mix, roadway design, congestion, terrain, and meteorology as the top AADT road segment in the area. These are good suggestions. Experiments should be designed to determine which variables most affect the ambient concentrations. References

18



1 2 3 4 5 6 7 8 9

10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44

Ayers, G.P.; Keywood, M.D.; Gillett, R.; Manins, P.C.; Malfroy, H.; Bardsley, T. (1998). Validation of passive diffusion samplers for SO2 and NO2. Atmos. Environ., 32(20): 3586-3592. Baik, J.J.; Kang, Y.S.; Kim, J.J. (2007). Modeling reactive pollutant dispersion in an urban street canyon. Atmos. Environ., 41(5): 934-949. Baik, J.J.; Kim, J.J. (2002). On the escape of pollutants from urban street canyons. Atmos. Environ., 36(3): 527-536. Bakeas, E.B.; Siskos, P.A. (2003). Dispersion of volatile hydrocarbons in urban street canyons. J. Air Waste Manage. Assoc., 53(4): 497-504. Baker, J.; Walker, H.L.; Cai, X.M. (2004). A study of the dispersion and transport of reactive pollutants in and above street canyons - a large eddy simulation. Atmos. Environ., 38(39): 6883-6892. Beckerman, B.; Jerrett, M.; Brook, J.R.; Verma, D.K.; Arain, M.A.; Finkelstein, M.M. (2008). Correlation of nitrogen dioxide with other traffic pollutants near a major expressway. Atmos. Environ., 42(2): 275-290. Berkowicz, R.; Ketzel, M.; Jensen, S.S.; Hvidberg, M.; Raaschou-Nielsen, O. (2008). Evaluation and application of OSPM for traffic pollution assessment for a large number of street locations. Environ. Modelling & Software, 23(3): 296-303. Boddy, J.W.D.; Smalley, R.J.; Dixon, N.S.; Tate, J.E.; Tomlin, A.S. (2005a). The spatial variability in concentrations of a traffic-related pollutant in two street canyons in York, UK - Part I: The influence of background winds. Atmos. Environ., 39(17): 3147-3161. Boddy, J.W.D.; Smalley, R.J.; Goodman, P.S.; Tate, J.E.; Bell, M.C.; Tomlin, A.S. (2005b). The spatial variability in concentrations of a traffic-related pollutant in two street canyons in York, UK - Part II: The influence of traffic characteristics. Atmos. Environ., 39(17): 3163-3176. Bowen, J.L.; Egami, R.T. (1994). Clark County Carbon Monoxide Hotspot Study. Report Number DRI 6460-684-4010.1F1; prepared by Desert Research Institute, Reno, NV, for Clark County Health District, Las Vegas, NV; Bowen, J.L.; Walsh, P.A.; Henry, R.C. (1993). Analysis of data from Lynwood Carbon Monoxide Study. Report Number DRI Document No. 685-2656.D2; prepared by Desert Research Institute, Reno, NV, for California Air Resources Board, Sacramento, CA; Brunet, J.; Pauly, A.; Varenne, C.; Lauron, B. (2008). On-board phthalocyanine gas sensor microsystem dedicated to the monitoring of oxidizing gases level in passenger compartments. Sensors and Actuators B-Chemical, 130(2): 908-916. Buccolieri, R.; Gromke, C.; Di Sabatino, S.; Ruck, B. (2009). Aerodynamic effects of trees on pollutant concentration in street canyons. Sci. Total Environ., 407(19): 5247-5256. Buccolieri, R.; Sandberg, M.; Di Sabatino, S. (2010). City breathability and its link to pollutant concentration distribution within urban-like geometries. Atmos. Environ., 44(15): 1894-1903. Bukowiechi, N.; Dommen, J.; Prevot, A.S.H.; Richter, R.; Weingartner, E.; Baltensperger, U. (2002). A mobile pollutant measurement laboratory-measuring gas phase and aerosol ambient concentrations with high spatial and temporal resolution. Atmos. Environ., 36(36-37): 5569-5579.

19



1 2 3 4 5 6 7 8 9

10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43

Bukowiecki, N.; Dommen, J.; Prevot, A.S.H.; Richter, R.; Weingartner, E.; Baltensperger, U. (2003). Fine and ultrafine particles in the Zurich (Switzerland) area measured with a mobile laboratory. Atmos. Chem. Phys. Discuss., 3: 2739-2782. Cadle, S.H.; Mulawa, P.; Groblicki, P.; Laroo, C.; Ragazzi, R.A.; Nelson, K.; Gallagher, G.; Zielinska, B. (2001). In-use light-duty gasoline vehicle particulate matter emissions on three driving cycles. Environ. Sci. Technol., 35(1): 26-32. Cai, X.M.; Barlow, J.F.; Belcher, S.E. (2008). Dispersion and transfer of passive scalars in and above street canyons - Large-eddy simulations. Atmos. Environ., 42(23): 5885-5895. Caton, F.; Britter, R.E.; Dalziel, S. (2003). Dispersion mechanisms in a street canyon. Atmos. Environ., 37(5): 693-702. Chakrabarty, S.; Sarkar, A.D.; Biswas, S. (2000). Validation of a kerbside urban air quality model for nitrogen dioxide. Bangalore University, Bangalore, pp. 25-29. Chan, A.T.; Au, W.T.W.; So, E.S.P. (2003). Strategic guidelines for street canyon geometry to achieve sustainable street air quality Part II: Multiple canopies and canyons. Atmos. Environ., 37(20): 2761-2772. Chan, L.Y.; Hung, W.T.; Qin, T. (1995). Assmessment of vehicular emission dispersion models applied in street canyons in Guangzhou, PRC. Environ. Int., 21: 39-46. Chan, S.H.; Zhu, J. (1999). Modelling of cold-start engine exhaust heat transfer and catalysts lightoff. Journal of the Institute of Energy, 72(490): 11-19. Chang, T.J.; Kao, H.M.; Wu, Y.T.; Huang, W.H. (2009). Transport mechanisms of coarse, fine, and very fine particulate matter in urban street canopies with different building layouts. J. Air Waste Manage. Assoc., 59(2): 196-206. Chase, R.E.; Duszkiewicz, G.J.; Jensen, T.E.; Lewis, D.; Schlaps, E.J.; Weibel, A.T.; Cadle, S.H.; Mulawa, P. (2000). Particle mass emission rates from current-technology, light-duty gasoline vehicles. J. Air Waste Manage. Assoc., 50(6): 930-935. Cheng, W.C.; Liu, C.H.; Leung, D.Y.C. (2008). Computational formulation for the evaluation of street canyon ventilation and pollutant removal performance. Atmos. Environ., 42(40): 9041-9051. Cheng, W.C.; Liu, C.H.; Leung, D.Y.C. (2009). On the correlation of air and pollutant exchange for street canyons in combined wind-buoyancy-driven flow. Atmos. Environ., 43(24): 3682-3690. Cheng, Y.H. (2008). Comparison of the TSI model 8520 and Grimm Series 1.108 portable aerosol instruments used to monitor particulate matter in an iron foundry. Journal of Occupational and Environmental Hygiene, 5(3): 157-168. Chow, J.C.; Bachmann, J.D.; Kinsman, J.D.; Legge, A.H.; Watson, J.G.; Hidy, G.M.; Pennell, W.R. (2010a). Multipollutant air quality management: Critical review discussion. J. Air Waste Manage. Assoc., 60(10): in press. Chow, J.C.; Engelbrecht, J.P.; Watson, J.G.; Wilson, W.E.; Frank, N.H.; Zhu, T. (2002). Designing monitoring networks to represent outdoor human exposure. Chemosphere, 49(9): 961-978. Chow, J.C.; Wang, X.L.; Kohl, S.D.; Gronstal, S.; Watson, J.G. (2010b). Heavy-duty diesel emissions in the Athabasca Oil Sands Region. In Proceedings, 103rd Annual

20



1 2 3 4 5 6 7 8 9

10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43

Meeting of the Air & Waste Management Association, Tropp, R. J., Legge, A. H., Eds.; Air & Waste Management Association: Pittsburgh, PA, 1-5. Chu, A.K.M.; Kwok, R.C.W.; Yu, K.N. (2005). Study of pollution dispersion in urban areas using Computational Fluid Dynamics (CFD) and Geographic Information System (GIS). Environ. Modelling & Software, 20(3): 273-277. Clarke, A.G.; Robertson, L.A.; Hamilton, R.S.; Gorbunov, B. (2004). A Lagrangian model of the evolution of the particulate size distribution of vehicular emissions. Sci. Total Environ., 334-335: 197-206. Cocker, D.R.; Shah, S.D.; Johnson, K.; Miller, J.W.; Norbeck, J.M. (2004a). Development and application of a mobile laboratory for measuring emissions from diesel engines 1. Regulated gaseous emissions. Environ. Sci. Technol., 38(7): 2182-2189. Cocker, D.R.; Shah, S.D.; Johnson, K.C.; Zhu, X.N.; Miller, J.W.; Norbeck, J.M. (2004b). Development and application of a mobile laboratory for measuring emissions from diesel engines. 2. Sampling for toxics and particulate matter. Environ. Sci. Technol., 38(24): 6809-6816. Cook, R.; Touma, J.S.; Fernandez, A.; Brzezinski, D.; Bailey, C.; Scarbro, C.; Thurman, J.; Strum, M.; Ensley, D.; Baldauf, R. (2007). Impact of underestimating the effects of cold temperature on motor vehicle start emissions of air toxics in the United States. J. Air Waste Manage. Assoc., 57: 1469-1479. Cotte, H.; Bessagnet, B.; Blondeau, C.; Mallet-Hubert, P.Y.; Momique, J.C.; Walter, C.; Boulanger, L.; Deleger, D.; Jouvenot, G.; Pain, C.; Rouveirolles, P. (2001). Cold-start emissions from petrol and diesel vehicles according to the emissions regulations (from Euro 92 to Euro 2000). Int. J. Vehicle Des., 27(1-4): 275-285. Crouse, D.L.; Goldberg, M.S.; Ross, N.A. (2009). A prediction-based approach to modelling temporal and spatial variability of traffic-related air pollution in Montreal, Canada. Atmos. Environ., 43(32): 5075-5084. Currie, J.F.; Essalik, A.; Marusic, J.C. (1999). Micromachined thin film solid state electrochemical CO2, NO2 and SO2 gas sensors. Sensors and Actuators B-Chemical, 59(2-3): 235-241. De Fouquet, C.; Gallois, D.; Perron, G. (2007). Geostatistical characterization of nitrogen dioxide concentration in an urban area - Part II: Time component of the estimation error. Atmos. Environ., 41(32): 6691-6700. Di Sabatino, S.; Buccolieri, R.; Pulvirenti, B.; Britter, R.E. (2008). Flow and pollutant dispersion in street canyons using FLUENT and ADMS-Urban. Environmental Modeling & Assessment, 13(3): 369-381. Dixon, N.S.; Boddy, J.W.D.; Smalley, R.J.; Tomlin, A.S. (2006). Evaluation of a turbulent flow and dispersion model in a typical street canyon in York, UK. Atmos. Environ., 40(5): 958-972. Do, J.S.; Chen, P. (2007). Amperometric sensor array for NOx, CO, O2 and SO2 detection. Sensors and Actuators B-Chemical, 122(1): 165-173. Dobre, A.; Arnold, S.J.; Smalley, R.J.; Boddy, J.W.D.; Barlow, J.F.; Tomlin, A.S.; Belcher, S.E. (2005). Flow field measurements in the proximity of an urban intersection in London, UK. Atmos. Environ., 39(26): 4647-4657.

21



1 2 3 4 5 6 7 8 9

10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43

Douglas, K.E.; Beaulieu, H.J. (1983). Field validation study of nitrogen dioxide passive samplers in a "diesel" haulage underground mine. J. Am. Ind. Hyg. Assoc., 44(10): 774-778. Durbin, T.D.; Johnson, K.; Cocker, D.R.; Miller, J.W.; Maldonado, H.; Shah, A.; Ensfield, C.; Weaver, C.; Akard, M.; Harvey, N.; Symon, J.; Lanni, T.; Bachalo, W.D.; Payne, G.; Smallwood, G.; Linke, M. (2007). Evaluation and comparison of portable emissions measurement systems and federal reference methods for emissions from a back-up generator and a diesel truck operated on a chassis dynamometer. Environ. Sci. Technol., 41(17): 6199-6204. Egashira, M.; Shimizu, Y.; Takao, Y.; Sako, S. (1996). Variations in I-V characteristics of oxide semiconductors induced by oxidizing gases. Sensors and Actuators B: Chemical, 35(1-3): 62-67. Eliasson, I.; Offerle, B.; Grimmond, C.S.B.; Lindqvist, S. (2006). Wind fields and turbulence statistics in an urban street canyon. Atmos. Environ., 40(1): 1-16. ISI:000234726200001. Faus-Kessler, T.; Kirchner, M.; Jakobi, G. (2008). Modelling the decay of concentrations of nitrogenous compounds with distance from roads. Atmos. Environ., 42(19): 4589-4600. Forleo, A.; Francioso, L.; Epifani, M.; Capone, S.; Taurino, A.M.; Siciliano, P. (2005). NO2-gas-sensing properties of mixed In2O3-SnO2 thin films. Thin Solid Films, 490(1): 68-73. Genikhovich, E.L.; Ziv, A.D.; Iakovleva, E.A.; Palmgren, F.; Berkowicz, R. (2005). Joint analysis of air pollution in street canyons in St. Petersburg and Copenhagen. Atmos. Environ., 39(15): 2747-2757. Gidhagen, L.; Johansson, C.; Langner, J.; Olivares, G. (2004a). Simulation of NOx and ultrafine particles in a street canyon in Stockholm, Sweden. Atmos. Environ., 38(14): 2029-2044. Gidhagen, L.; Johansson, C.; Omstedt, G.; Langner, J.; Olivares, G. (2004b). Model simulations of NOx and ultrafine particles close to a Swedish highway. Environ. Sci. Technol., 38(24): 6730-6740. Gilbert, N.L.; Woodhouse, S.; Stieb, D.M.; Brook, J.R. (2003). Ambient nitrogen dioxide and distance from a major highway. Sci. Total Environ., 312: 43-46. Gokhale, S.B.; Rebours, A.; Pavageau, M. (2005). The performance evaluation of WinOSPM model for urban street canyons of Nantes in France. Environ. Mon. Assess., 100(1-3): 153-176. Gonzales, M.; Qualls, C.; Hudgens, E.; Neas, L. (2005). Characterization of a spatial gradient of nitrogen dioxide across a United States-Mexico border city during winter. Sci. Total Environ., 337(1-3): 163-173. Grawe, D.; Cai, X.M.; Harrison, R.M. (2007). Large eddy simulation of shading effects on NO2 and O-3 concentrations within an idealised street canyon. Atmos. Environ., 41(34): 7304-7314. Greenbaum, D.; Shaikh, R. (2010). First steps toward multipollutant science for air quality decisions. Epidemiology, 21(2): 195-197.

22



1 2 3 4 5 6 7 8 9

10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Gromke, C.; Buccolieri, R.; Di Sabatino, S.; Ruck, B. (2008). Dispersion study in a street canyon with tree planting by means of wind tunnel and numerical investigations - Evaluation of CFD data with experimental data. Atmos. Environ., 42(37): 8640-8650. Gullett, B.K.; Touati, A.; Oudejans, L.; Ryan, S.P. (2006). Real-time emission characterization of organic air toxic pollutants during steady state and transient operation of a medium duty diesel engine. Atmos. Environ., 40(22): 4037-4047. Gurlo, A.; BGrsan, N.; Ivanovskaya, M.; Weimar, U.; G÷pel, W. (1998). In2O3 and MoO3-In2O3 thin film semiconductor sensors: Interaction with NO2 and O3. Sensors and Actuators B: Chemical, 47(1-3): 92-99. Hang, J.; Sandberg, M.; Li, Y.G. (2009). Effect of urban morphology on wind condition in idealized city models. Atmos. Environ., 43(4): 869-878. Hansen, A.D.A.; Mocnik, G. (2010). The "Micro" Aethalometer(R) - An enabling technology for new applications in the measurement of aerosol black carbon. In Proceedings, Leapfrogging Opportunities for Air Quality Improvement, Chow, J. C., Watson, J. G., Cao, J. J., Eds.; Air & Waste Management Association: Pittsburgh, PA, 984-989. Hauser, C.D.; Battle, P.; Mace, N. (2009). Adjusted blank correction method for UV-vis spectroscopic analysis of PTIO-coated filters used in nitrogen oxide passive samplers. Atmos. Environ., 43(10): 1823-1826. Heal, M.R.; Cape, J.N. (1997). A numerical evaluation of chemical interferences in the measurement of ambient nitrogen dioxide by passive diffusion samplers. Atmos. Environ., 31(13): 1911-1923. Heal, M.R.; Kirby, C.; Cape, J.N. (2000). Systematic biases in measurement of urban nitrogen dioxide using passive diffusion samplers. Environ. Mon. Assess., 62(1): 39-54. Heal, M.R.; O'Donoghue, M.A.; Cape, J.N. (1999). Overestimation of urban nitrogen dioxide by passive diffusion tubes: A comparative exposure and model study. Atmos. Environ., 33(4): 513-524. Heim, M.; Mullins, B.J.; Umhauer, H.; Kasper, G. (2008). Performance evaluation of three optical particle counters with an efficient "multimodal" calibration method. J. Aerosol Sci., 39(12): 1019-1031. Henderson, S.B.; Beckerman, B.; Jerrett, M.; Brauer, M. (2007). Application of land use regression to estimate long-term concentrations of traffic-related nitrogen oxides and fine particulate matter. Environ. Sci. Technol., 41(7): 2422-2428. Herndon, S.C.; Shorter, J.H.; Zahniser, M.S.; Wormhoudt, J.; Nelson, D.D.; Demerjian, K.L.; Kolb, C.E. (2005). Real-time measurements of SO2, H2CO, and CH4 emissions from in-use curbside passenger buses in New York City using a chase vehicle. Environ. Sci. Technol., 39(20): 7984-7990. Hidy, G.M.; Pennell, W.R. (2010). Multipollutant air quality management: A critical review. J. Air Waste Manage. Assoc., 60(6): 645-674. Huai, T.; Durbin, T.D.; Miller, J.W.; Norbeck, J.M. (2004). Estimates of the emission rates of nitrous oxide from light-duty vehicles using different chassis dynamometer test cycles. Atmos. Environ., 38(38): 6621-6629.

23



1 2 3 4 5 6 7 8 9

10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43

Isakov, V.; Touma, J.S.; Khlystov, A. (2007). A method of assessing air toxics concentrations in urban areas using mobile platform measurements. J. Air Waste Manage. Assoc., 57(11): 1286-1295. Joumard, R.; Andre, M. (1990). Cold start emissions of traffic. Sci. Total Environ., 93: 175-182. Joumard, R.; Andre, M.; Vidon, R.; Tassel, P.; Pruvost, C. (2000). Influence of driving cycles on unit emissions from passenger cars. Atmos. Environ., 34(27): 4621-4628. Kang, Y.S.; Baik, J.J.; Kim, J.J. (2008). Further studies of flow and reactive pollutant dispersion in a street canyon with bottom heating. Atmos. Environ., 42(20): 4964-4975. Karner, A.A.; Eisinger, D.S.; Niemeier, D.A. (2010). Near-roadway air quality: Synthesizing the findings from real-world data. Environ. Sci. Technol., 44(14): 5334-5344. Kassomenos, P.A.; Karakitsios, S.P.; Pilidis, G.A. (2004). A simple semi-empirical approach to modeling benzene concentrations in a street canyon. Atmos. Environ., 38(35): 6073-6078. ISI:000224654100018. Kean, A.J.; Sawyer, R.F.; Harley, R.A. (2000). A fuel-based assessment of off-road diesel engine emissions. J. Air Waste Manage. Assoc., 50(11): 1929-1939. Kim, J.J.; Baik, J.J. (2004). A numerical study of the effects of ambient wind direction on flow and dispersion in urban street canyons using the RNG k-epsilon turbulence model. Atmos. Environ., 38(19): 3039-3048. Kittelson, D.B.; Watts, W.F.; Johnson, J.P.; Remerowki, M.L.; Ische, E.E.; Oberdörster, G.; Gelein, R.A.; Elder, A.; Hopke, P.K.; Kim, E.; Zhao, W.; Zhou, L.; Jeong, C.H. (2004). On-road exposure to highway aerosols. 1. Aerosol and gas measurements. Inhal. Toxicol., 16(Suppl. 1): 31-39. Kittelson, D.B.; Watts, W.F.; Johnson, J.P.; Schauer, J.J.; Lawson, D.R. (2006). On-road and laboratory evaluation of combustion aerosols - Part 2: Summary of spark ignition engine results. J. Aerosol Sci., 37(8): 931-949. Kok, G.L.; Baumgardner, D. (2010). Emerging technologies for measurements of black carbon. In Proceedings, Leapfrogging Opportunities for Air Quality Improvement, Chow, J. C., Watson, J. G., Cao, J. J., Eds.; Air & Waste Management Association: Pittsburgh, PA, 996. Kondo, H.; Asahi, K.; Tomizuka, T.; Suzuki, M. (2006). Numerical analysis of diffusion around a suspended expressway by a multi-scale CFD model. Atmos. Environ., 40(16): 2852-2859. Kondo, H.; Tomizuka, T. (2009). A numerical experiment of roadside diffusion under traffic-produced flow and turbulence. Atmos. Environ., 43(27): 4137-4147. Korin, E.; Reshef, R.; Tshernichovesky, D.; Sher, E. (1999). Reducing cold-start emission from internal combustion engines by means of a catalytic converter embedded in a phase-change material. Proceedings of the Institution of Mechanical Engineers Part D-Journal of Automobile Engineering, 213(D6): 575-583. Krochmal, D.; Gorski, L. (1991). Determination of nitrogen dioxide in ambient air by use of a passive sampling technique and triethanolamine as absorbent. Environ. Sci. Technol., 25(3): 531-535.

24



1 2 3 4 5 6 7 8 9

10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44

Kumar, P.; Fennell, P.; Langley, D.; Britter, R. (2008). Pseudo-simultaneous measurements for the vertical variation of coarse, fine and ultrafine particles in an urban street canyon. Atmos. Environ., 42(18): 4304-4319. Kumar, P.; Garmory, A.; Ketzel, M.; Berkowicz, R.; Britter, R. (2009). Comparative study of measured and modelled number concentrations of nanoparticles in an urban street canyon. Atmos. Environ., 43(4): 949-958. Lam, G.C.K.; Leung, D.Y.C.; Niewiadomski, M.; Pang, S.W.; Lee, A.W.F.; Louie, P.K.K. (2008). Street level concentrations of nitrogen dioxide and suspended particulate matter in Hong Kong. Atmos. Environ., 33: 1-11. Lenaers, G. (1996). On-board real life emission measurements on a 3 way catalyst gasoline car in motor way-, rural- and city traffic and on two Euro-1 diesel city buses. Sci. Total Environ., 190: 139-147. Li, X.X.; Liu, C.H.; Leung, D.Y.C. (2005). Development of a k-epsilon model for the determination of air exchange rates for street canyons. Atmos. Environ., 39(38): 7285-7296. Li, X.X.; Liu, C.H.; Leung, D.Y.C. (2009). Numerical investigation of pollutant transport characteristics inside deep urban street canyons. Atmos. Environ., 43(15): 2410-2418. Li, X.X.; Liu, C.H.; Leung, D.Y.C.; Lam, K.M. (2006). Recent progress in CFD modelling of wind field and pollutant transport in street canyons. Atmos. Environ., 40(29): 5640-5658. Linnainmaa, M.; Laitinen, J.; Leskinen, A.; Sippula, O.; Kalliokoski, P. (2008). Laboratory and field testing of sampling methods for inhalable and respirable dust. Journal of Occupational and Environmental Hygiene, 5(1): 28-35. Liu, C.H.; Leung, D.Y.C. (2008). Numerical study on the ozone formation inside street canyons using a chemistry box model. Journal of Environmental Sciences-China, 20(7): 832-837. Longley, I.D. (2004). Corrections to a description of turbulence in a trafficked street canyon in Manchester. Atmos. Environ., 38(27): 4589-4592. Longley, I.D.; Gallager, M.W.; Dorsey, J.R.; Flynn, M.; Allan, J.D.; Alfarra, M.R.; Inglis, D. (2003). A case study of aerosol (4.5 nm<Dp<10µm) number and mass size distribution measurements in a busy street canyon in Manchester, UK. Atmos. Environ., 37(12): 1563-1571. Longley, I.D.; Gallagher, M.W.; Dorsey, J.R.; Flynn, M.; Barlow, J.F. (2004). Short-term measurements of airflow and turbulence in two street canyons in Manchester. Atmos. Environ., 38(1): 69-79. Lough, G.C.; Schauer, J.J.; Lonneman, W.A.; Allen, M.K. (2005). Summer and winter nonmethane hydrocarbon emissions from on-road motor vehicles in the midwestern United States. J. Air Waste Manage. Assoc., 55(5): 629-646. Ludykar, D.; Westerholm, R.; Almén, J. (1999). Cold start emissions at +22, -7 and -20°C ambient temperatures from a three way catalyst (TWC) car: Regulated and unregulated exhaust components. Sci. Total Environ., 235(1-3): 65-69. Maricq, M.M.; Podsiadlik, D.H.; Chase, R.E. (1999). Examination of the size-resolved and transient nature of motor vehicle particle emissions. Environ. Sci. Technol., 33(10): 1618-1626.

25



1 2 3 4 5 6 7 8 9

10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Mathis, U.; Mohr, M.; Forss, A.M. (2005). Comprehensive particle characterization of modern gasoline and diesel passenger cars at low ambient temperatures. Atmos. Environ., 39(1): 107-117. Mauderly, J.L.; Burnett, R.T.; Castillejos, M.; Ozkaynak, H.; Samet, J.M.; Stieb, D.M.; Vedal, S.; Wyzga, R.E. (2010). Commentary: Is the air pollution health research community prepared to support a multipollutant air quality management framework? Inhal. Toxicol., 22(S1): 1-19. McNabola, A.; Broderick, B.M.; Gill, L.W. (2009). A numerical investigation of the impact of low boundary walls on pedestrian exposure to air pollutants in urban street canyons. Sci. Total Environ., 407(2): 760-769. Molina, M.J. (1996). Polar ozone depletion (Nobel lecture). Angewandte Chemie-International Edition in English, 35(16): 1778-1785. Morawska, L.; Ristovski, Z.D.; Johnson, G.R.; Jayaratne, E.R.; Mengersen, K. (2007). Novel method for on-road emission factor measurements using a plume capture trailer. Environ. Sci. Technol., 41(2): 574-579. Moussiopoulos, N.; Barmpas, P.; Ossanlis, I.; Bartzis, J. (2008). Comparison of numerical and experimental results for the evaluation of the depollution effectiveness of photocatalytic coverings in street canyons. Environmental Modeling & Assessment, 13(3): 357-368. Mukerjee, S.; Smith, L.A.; Norris, G.A.; Morandi, M.T.; Gonzales, M.; Noble, C.A.; Neas, L.M.; Özkaynak, A.H. (2004). Field method comparison between passive air samplers and continuous monitors for VOCs and NO2 in El Paso, Texas. J. Air Waste Manage. Assoc., 54(3): 307-319. Murena, F.; Favale, G.; Vardoulakis, S.; Solazzo, E. (2009). Modelling dispersion of traffic pollution in a deep street canyon: Application of CFD and operational models. Atmos. Environ., 43(14): 2303-2311. Murena, F.; Garofalo, N.; Favale, G. (2008). Monitoring CO concentration at leeward and windward sides in a deep street canyon. Atmos. Environ., 42(35): 8204-8210. Murena, F.; Vorraro, F. (2003). Vertical gradients of benzene concentration in a deep street canyon in the urban area of Naples. Atmos. Environ., 37(35): 4853-4859. Nash, D.G.; Leith, D. (2010). Use of passive diffusion tubes to monitor air pollutants. J. Air Waste Manage. Assoc., 60(2): 204-209. Ning, Z.; Cheung, C.S.; Lu, Y.; Liu, M.A.; Hung, W.T. (2005). Experimental and numerical study of the dispersion of motor vehicle pollutants under idle condition. Atmos. Environ., 39(40): 7880-7893. Norris, G.; Larson, T. (1999). Spatial and temporal measurements of NO2 in an urban area using continuous mobile monitoring and passive samplers. J. Expo. Anal. Environ. Epidemiol., 9(6): 586-593. Nussbaum, N.J.; Zhu, D.; Kuhns, H.D.; Mazzoleni, C.; Chang, M.-C.O.; Moosmüller, H.; Watson, J.G. (2009). The In-Plume Emissions Test-Stand: A novel instrument platform for the real-time characterization of combustion emissions. J. Air Waste Manage. Assoc., 59(12): 1437-1445.

26



1 2 3 4 5 6 7 8 9

10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44

Oettl, D.; Hausberger, S.; Rexeis, M.; Sturm, P.J. (2006). Simulation of traffic induced NOx concentrations near the A12 highway in Austria. Atmos. Environ., 40(31): 6043-6052. Oto, K.; Shinobe, A.; Manabe, M.; Kakuuchi, H.; Yoshida, Y.; Nakahara, T. (2001). New semiconductor type gas sensor for air quality control in automobile cabin. Sensors and Actuators B: Chemical, 77(1-2): 525-528. Ozden, O.; Dogeroglu, T. (2008). Field evaluation of a tailor-made new passive sampler for the determination of NO2 levels in ambient air. Environ. Mon. Assess., 142(1-3): 243-253. Parra, M.A.; Elustondo, D.; Bermejo, R.; Santamaria, J.M. (2009). Ambient air levels of volatile organic compounds (VOC) and nitrogen dioxide (NO2) in a medium size city in Northern Spain. Sci. Total Environ., 407(3): 999-1009. Piechocki-Minguy, A.; Plaisance, H.; Schadkowski, C.; Sagnier, I.; Saison, J.Y.; Galloo, J.C.; Guillermo, R. (2006). A case study of personal exposure to nitrogen dioxide using a new high sensitive diffusive sampler. Sci. Total Environ., 366(1): 55-64. Pirjola, L.; Parviainen, H.; Hussein, T.; Valli, A.; Hameri, K.; Aaalto, P.; Virtanen, A.; Keskinen, J.; Pakkanen, T.A.; Makela, T.; Hillamo, R.E. (2004). "Sniffer" - a novel tool for chasing vehicles and measuring traffic pollutants. Atmos. Environ., 38(22): 3625-3635. Plaisance, H.; Plechocki-Minguy, A.; Garcia-Fouque, S.; Galloo, J.C. (2004). Influence of meteorological factors on the NO2 measurements by passive diffusion tube. Atmos. Environ., 38(4): 573-580. Pohjola, M.; Pirjola, L.; Kukkonen, J.; Kulmala, M. (2003). Modelling of the influence of aerosol processes for the dispersion of vehicular exhaust plumes in street environment. Atmos. Environ., 37(3): 339-351. Pornet, P.; Beaubestre, C.; Courtois, Y.; Festy, B.; Ing, H.; Lopez, B.; Marduel, J.L.; Beurdouche, P.; Chevrier, M.; Hublin, M.; Jouvenot, G.; Laliere, M.; Tarriere, C. (1995). Impact of driving conditions on efficiency of catalytic converters in diesel and gas-powered vehicles. Sci. Total Environ., 169(1-3): 321-329. Prajapati, S.K.; Tripathi, B.D.; Pathak, V. (2009). Distribution of vehicular pollutants in street canyons of Varanasi, India: a different case. Environ. Mon. Assess., 148(1-4): 167-172. Rava, M.; Verlato, G.; Bono, R.; Ponzio, M.; Sartori, S.; Blengio, G.; Kuenzli, N.; Heinrich, J.; Gotschi, T.; de Marco, R. (2007). A predictive model for the home outdoor exposure to nitrogen dioxide. Sci. Total Environ., 384: 163-170. Ristimaki, J.; Keskinen, J.; Virtanen, A.; Maricq, M.; Aakko, P. (2005). Cold temperature PM emissions measurement: Method evaluation and application to light duty vehicles. Environ. Sci. Technol., 39(24): 9424-9430. Rodden, J.B.; Green, N.J.; Messina, A.D.; Bullin, J.A. (1982). Comparison of roadway pollutant dispersion models using the Texas data. J. Air Poll. Control Assoc., 32(12): 1226-1228. Sahlodin, A.M.; Sotudeh-Gharebagh, R.; Zhu, Y.F. (2007). Modeling of dispersion near roadways based on the vehicle-induced turbulence concept. Atmos. Environ., 41(1): 92-102.

27



1 2 3 4 5 6 7 8 9

10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43

Santiago, J.L.; Martin, F. (2008). SLP-2D: A new Lagrangian particle model to simulate pollutant dispersion in street canyons. Atmos. Environ., 42(17): 3927-3936. Sawyer, R.F.; Harley, R.A.; Cadle, S.H.; Norbeck, J.M.; Slott, R.S.; Bravo, H.A. (2000). Mobile sources critical review: 1998 NARSTO assessment. Atmos. Environ., 34(12-14): 2161-2181. Scaperdas, A.; Colvile, R.N. (1999). Assessing the representativeness of monitoring data from an urban intersection site in central London, UK. Atmos. Environ., 33(4): 661-674. Schauer, J.J.; Christensen, C.G.; Kittelson, D.B.; Johnson, J.P.; Watts, W.F. (2008). Impact of ambient temperatures and driving conditions on the chemical composition of particulate matter emissions from non-smoking gasoline-powered motor vehicles. Aerosol Sci. Technol., 42(3): 210-223. Seigneur, C. (2009). Current understanding of ultrafine particulate matter emitted from mobile sources. J. Air Waste Manage. Assoc., 59(1): 3-17. Sekine, Y.; Watts, S.F.; Rendell, A.; Butsugan, M. (2008). Development of highly sensitive passive sampler for nitrogen dioxide using porous polyethylene membrane filter as turbulence limiting diffuser. Atmos. Environ., 42(18): 4079-4088. Shooter, D.; Brimblecombe, P.; Shooter, J.; Lowe, D.; Day, P.J.; Du, S. (1997). Some characteristics and applications of nitrogen dioxide passive samplers. Environmental Technology, 18(3): 243-254. Singer, B.C.; Kirchstetter, T.W.; Harley, R.A.; Kendall, G.R.; Hesson, J.M. (1999). A fuel-based approach to estimating motor vehicle cold-start emissions. J. Air Waste Manage. Assoc., 49(2): 125-135. Sitnikov, N.M.; Sokolov, A.O.; Ravengnani, F.; Yushkov, V.A.; Ulanovsky, A.E. (2005). A chemiluminescent balloon-type nitrogen dioxide meter for tropospheric and stratospheric investigations (NaDA). Instruments and Experimental Techniques, 48(3): 400-405. Smichowski, P.; Gomez, D.; Frazzoli, C.; Caroji, S. (2008). Traffic-related elements in airborne particulate matter. Applied Spectroscopy Reviews, 43(1): 22-48. So, E.S.P.; Chan, A.T.Y.; Wong, A.Y.T. (2005). Large-eddy simulations of wind flow and pollutant dispersion in a street canyon. Atmos. Environ., 39(20): 3573-3582. Solazzo, E.; Vardoulakis, S.; Cai, X.M. (2007). Evaluation of traffic-producing turbulence schemes within operational street pollution models using roadside measurements. Atmos. Environ., 41(26): 5357-5370. Stein, A.F.; Toselli, B.M. (1996). Street level air pollution in Cordoba City, Argentina. Atmos. Environ., 30(20): 3491-3495. Talazac, L.; Barbarin, F.; Varenne, C.; Cuminal, Y. (2001). Highly NO2 sensitive pseudo Schottky barrier diodes on p-type InP with improved electrical characteristics. Sensors and Actuators B: Chemical, 77(1-2): 447-454. Tay, B.K.; McFiggans, G.B.; Jones, D.P.; Gallagher, M.W.; Martin, C.; Watkins, P.; Harrison, R.M. (2010). Linking urban aerosol fluxes in street canyons to larger scale emissions. Atmos. Chem. Phys., 10(5): 2475-2490. Tsai, M.Y.; Chen, K.S. (2004). Measurements and three-dimensional modeling of air pollutant dispersion in an Urban Street Canyon. Atmos. Environ., 38(35): 5911-5924.

28



1 2 3 4 5

Tsai, M.Y.; Chen, K.S.; Wu, C.H. (2005). Three-dimensional modeling of air flow and pollutant dispersion in an urban street canyon with thermal effects. J. Air Waste Manage. Assoc., 55(8): 1178-1189. U.S.EPA (1997). Guidance for network design and optimum site exposure for PM2.5 and PM10. Report Number EPA-454/R-99-022; prepared by U.S. Environmental Protection Agency, Research Triangle Park, NC, http://www.epa.gov/ttn/amtic/pmstg.html 6

7 8 9

10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43

U.S.EPA (1998). Guideline of ozone monitoring site selection. Report Number EPA-454/R-98-002, August 1998; prepared by U.S. EPA, OAQPS, Research Triangle Park, Ulanovsky, A.E.; Yushkov, V.A.; Sitnikov, N.M.; Ravengnani, F. (2001). The FOZAN-II fast-response chemiluminescent airborne ozone analyzer. Instruments and Experimental Techniques, 44: 249-256. Vallejos, S.; Khatko, V.; Aguir, K.; Ngo, K.A.; Calderer, J.; Gracia, I.; Cane, C.; Llobet, E.; Correig, X. (2007). Ozone monitoring by micro-machined sensors with WO3 sensing films. Sensors and Actuators B-Chemical, 126(2): 573-578. Van Reeuwijk, H.; Fischer, P.H.; Harssema, H.; Briggs, D.J. (1998). Field comparison of two NO2 passive samplers to assess spatial variation. Environ. Mon. Assess., 50(1): 37-51. Vardoulakis, S.; Fisher, B.E.A.; Gonzalez-Flesca, N.; Pericleous, K. (2002). Model sensitivity and uncertainty analysis using roadside air quality measurements. Atmos. Environ., 36(13): 2121-2134. Vardoulakis, S.; Fisher, B.E.A.; Pericleous, K.; Gonzalez-Flesca, N. (2003). Modelling air quality in street canyons: A review. Atmos. Environ., 37(2): 155-182. Vardoulakis, S.; Lumbreras, J.; Solazzo, E. (2009). Comparative evaluation of nitrogen oxides and ozone passive diffusion tubes for exposure studies. Atmos. Environ., 43(16): 2509-2517. Venegas, L.E.; Mazzeo, N.A. (2000). Carbon monoxide concentration in a street canyon of Buenos Aires City (Argentina). Environ. Mon. Assess., 65(1-2): 417-424. Venkatram, A.; Isakov, V.; Thoma, E.D.; Baldauf, R.W. (2007). Analysis of air quality data near roadways using a dispersion model. Atmos. Environ., 41(40): 9481-9497. Voigtlander, J.; Tuch, T.; Birmili, W.; Wiedensohler, A. (2006). Correlation between traffic density and particle size distribution in a street canyon and the dependence on wind direction. Atmos. Chem. Phys., 6: 4275-4286. Wang, J.S.; Chan, T.L.; Cheung, C.S.; Leung, C.W.; Hung, W.T. (2006). Three-dimensional pollutant concentration dispersion of a vehicular exhaust plume in the real atmosphere. Atmos. Environ., 40(3): 484-497. Wang, X.L.; Chancellor, G.; Evenstad, J.; Farnsworth, J.E.; Hase, A.; Olson, G.M.; Sreenath, A.; Agarwal, J.K. (2009). A novel optical instrument for estimating size segregated aerosol mass concentration in real time. Aerosol Sci. Technol., 43: 939-950. Wang, Y.J.; Zhang, K.M. (2009). Modeling near-road air quality using a computational fluid dynamics model, CFD-VIT-RIT. Environ. Sci. Technol., 43(20): 7778-7783. Weilenmann, M.; Favez, J.Y.; Alvarez, R. (2009). Cold-start emissions of modern passenger cars at different low ambient temperatures and their evolution over vehicle legislation categories. Atmos. Environ., 43(15): 2419-2429.

29

http://www.epa.gov/ttn/amtic/pmstg.html



1 2 3 4 5 6 7 8 9

10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39

Weilenmann, M.; Soltic, P.; Saxer, C.; Forss, A.M.; Heeb, N. (2005). Regulated and nonregulated diesel and gasoline cold start emissions at different temperatures. Atmos. Environ., 39(13): 2433-2441. Westerholm, R.; Christensen, A.; Rosen, A. (1996). Regulated and unregulated exhaust emissions from two three-way catalyst equipped gasoline fuelled vehicles. Atmos. Environ., 30(20): 3529-3536. Xie, S.D.; Zhang, Y.H.; Li, Q.; Tang, X.Y. (2003). Spatial distribution of traffic-related pollutant concentrations in street canyons. Atmos. Environ., 37(23): 3213-3224. Xie, X.M.; Huang, Z.; Wang, J.S. (2005). Impact of building configuration on air quality in street canyon. Atmos. Environ., 39(25): 4519-4530. Xie, X.M.; Liu, C.H.; Leung, D.Y.C. (2007). Impact of building facades and ground heating on wind flow and pollutant transport in street canyons. Atmos. Environ., 41: 9030-9049. Xie, X.M.; Liu, C.H.; Leung, D.Y.C.; Leung, M.K.H. (2006). Characteristics of air exchange in a street canyon with ground heating. Atmos. Environ., 40(33): 6396-6409. Yassin, M.F.; Kellnerova, R.; Janour, Z. (2008). Impact of street intersections on air quality in an urban environment. Atmos. Environ., 42(20): 4948-4963. Yassin, M.F.; Kellnerova, R.; Janour, Z. (2009). Numerical simulation on pollutant dispersion from vehicle exhaust in street configurations. Environ. Mon. Assess., 156(1-4): 257-273. Yim, S.H.L.; Fung, J.C.H.; Lau, A.K.H.; Kot, S.C. (2009). Air ventilation impacts of the "wall effect" resulting from the alignment of high-rise buildings. Atmos. Environ., 43(32): 4982-4994. Zhang, K.M.; Wexler, A.S. (2004). Evolution of particle number distribution near roadways Part I: Analysis of aerosol dynamics and its implications for engine emission measurement. Atmos. Environ., 38(38): 6643-6653. Zhou, Y.; Levy, J.I. (2007). Factors influencing the spatial extent of mobile source air pollution impacts: a meta-analysis. Bmc Public Health, 7(1): 89. Zhou, Y.; Levy, J.I. (2008). The impact of urban street canyons on population exposure to traffic-related primary pollutants. Atmos. Environ., 42(13): 3087-3098. Zhu, Y.F.; Hinds, W.C. (2005). Predicting particle number concentrations near a highway based on vertical concentration profile. Atmos. Environ., 39(8): 1557-1566. Zou, X.D.; Shen, Z.M.; Yuan, T.; Yin, S.; Cai, J.P.; Chen, L.P.; Wang, W.H. (2006). Shifted power-law relationship between NO2 concentration and the distance from a highway: A new dispersion model based on the wind profile model. Atmos. Environ., 40(40): 8068-8073.

30



1 2 3 4 5 6 7 8 9

10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44

Mr. Bart Croes It’s not clear what EPA is trying to accomplish with its proposed near road monitoring program. If it’s to determine compliance with the new 1-hour national ambient air quality standard for nitrogen dioxide (NO2), then it seems that the primary focus should be on monitoring population centers rather than siting monitors near roadways. For example, how will nonattainment boundaries be established for these microscale environments? California has had a one-hour standard for NO2 for several decades and studied peak levels and trends throughout the State. Although future NO2 levels will be increasingly driven by direct emissions from heavy-duty diesel vehicles, current peak locations in California are in intermediate downwind areas (e.g., eastern border of Los Angeles County) where photochemical conversion (and not just immediate ozone titration) has taken place, not in the source areas. If the purpose is to conduct multi-pollutant monitoring to help inform exposure and health studies, then linkages with these types of research studies appears to be missing from the documents. The timeframe for establishing a national network (by January 1, 2013) seems much too rushed to allow for full availability and analysis of results from the near road monitoring pilot study.

Charge Questions: 1. The accompanying draft guidance document outline provides an initial thought of

the major topics required in the near-road monitoring guidance that will aid state monitoring agencies in the identification and implementation of NO2 near road monitoring sites from a multi-pollutant perspective. Please comment on the overall content of the recommended topics in the draft outline. Please provide suggestions on any missing subjects that should be included in the guidance document and any unnecessary topics that are currently listed in the attached draft, if applicable.

It’s not clear what EPA is trying to accomplish with its proposed near road monitoring program. The goals should be clearly delineated in order to comment on this brief outline. It would also be helpful to see the target number of pages for each section and the extent of the literature review. As the outline points out, the available literature is extensive (see the Health Effects Institute special report on Traffic-Related Air Pollution: A Critical Review of the Literature on Emissions, Exposure, and Health Effects) and a conducting a synthesis of existing near road and saturation monitoring datasets would be a useful addition. 2. EPA and NACAA envision the near-road guidance document to be written from a

multi-pollutant perspective. What pollutants and sub-species does the subcommittee believe should be included for consideration and discussion in the near-road monitoring guidance? Some potential species for consideration include

31



1 2 3 4 5 6 7 8 9

10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44

Again, some clarity is needed on what EPA is trying to accomplish with its proposed near road monitoring program. If it’s to determine compliance with existing air quality standards, then only NO2 and CO are all that is necessary, as the roadway environment as not the peak location for PM2.5 and PM10 (because secondary PM is so important in determining peak locations). If the purpose is to follow motor vehicle criteria pollutant and air toxic emission trends, or to conduct multi-pollutant monitoring to help inform exposure and health studies, then NOX, PM, black carbon, and the air toxics should also be included.

3. Identifying Candidate Near-road Site Areas a. Annual Average Daily Traffic (AADT) & Fleet Mix – To consider fleet mix with

regard to NO2, an idea is to encourage states that have fleet mix information to take an approach that uses average, fleet-wide grams per mile emissions estimates (one for light duty vehicles and one for heavy duty vehicles), combined with AADT information to further weight which road segments in an area may be more conducive to produce peak pollutant concentrations. EPA would use the latest emission factor information to aid such a calculation. Given the variability in emission rates from on-road vehicles based on vehicle technology, fuel, speed, environmental conditions, etc., does the subcommittee believe this approach is an appropriate way to “consider” fleet mix in near-road site selection or is a more refined inventory and modeling analysis required? The highest CO sites in California are the ones with the highest AADT for light-duty vehicles and oldest fleet, and the peak one-hour levels occur during winter. In the past, the highest NO2 levels were the same sites and season because of the reaction 2 NO + O2 2 NO2, which is important at low temperatures and NOX > 1 ppm. But future NO2 levels will be dominated by heavy-duty diesel vehicles (HDDV) and near-source mixing with ozone. Thus, a key factor in roadside NO2 levels is the fraction of heavy-duty diesel vehicles (HDDV) that are equipped (2007-2009 model year) or retrofitted with a catalyzed diesel particulate filter. In California, while most diesel vehicles emit 5% NO2 (from Caldecott Tunnel data), this can increase to 17% for retrofits and 45% for 2007-2009 model year (although absolute NOX, and consequently NO2 is lowered by 50%). In other states that do not have California’s NO2 limit for retrofits (or just went into place in 2009), the NO2 fraction can be much higher, although admittedly the number of retrofits is a much lower in the rest of the U.S. Since existing EPA emission models predict NOX and not NO2, this capability would need to be added.

32



1

3 4 5 6 7 8 9

10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44

b. AADT & Fleet Mix – Further, should the suggested approach above in question 2

3a to consider fleet mix via the use of average, fleet-wide emission factors, or the use of inventory and modeling analysis, take into account mobile source controls that are “on the books” but have not yet been fully realized due to fleet turnover? If so, how far out into the future should states consider their effects? NOX and NO2 emissions will continue to decrease as the 2010 model year HDDVs (with 90% NOX control over pre-2007 vehicles) become a bigger fraction of the fleet. Since these decreases will take place somewhat uniformly across the nation, the purpose of this question isn’t clear. Is the intent to forecast emission trends for States to get out of monitoring requirements?

c. Roadway Design – Studies suggest and support the concept that roadway design

influences pollutant dispersion near the road. The EPA suggests establishing sites at-grade with the road, without any nearby obstructions to air flow; however, the Agency recognizes that this might not always be feasible. Does the subcommittee agree with this recommendation for locating sites at-grade with no obstructions? What priority should be placed on this factor within the guidance, given the need for flexibility in identifying appropriate site locations? Obstructions between the roadways and the monitors should be minimized, but shouldn’t proximity to where people live be a more important consideration? A more practical consideration in site selection is the cost of leases. For an ARB freeway study in Lodi, rent started out at $750/month for both sites. Once the landowners realized that their proximity to the freeway was worthwhile to ARB they increased the rent to $1500/month.

d. Congestion Patterns – The congestion of a roadway can be estimated by the metric “Level of Service” (LOS). LOS uses a letter grade from A to F to identify a roadway’s performance, with “A” the best conditions where traffic flows at or above the posted speed limit and all motorists have complete mobility between lanes to “F” the worst congestion where travel time cannot be predicted and generally traffic demand exceeds the facility’s capacity. Since motor vehicles generally emit more pollutants during congestion operations (although noting that NOX and select other pollutant emissions can also increase with increasing speed), how important a parameter should LOS be in the determination of appropriate near-road monitoring sites? Does the subcommittee have a view on how reliable LOS estimates are across the country? As NOX and VOC emissions are continuously reduced over time, ambient oxidant levels will also go down, which means that titration of NO by ozone to form NO2 will become less important in comparison to direct NO2 emissions from new and retrofitted diesel trucks. NO2 emissions are highest during cruise mode, not

33



1 2 3 4

6 7 8 9

10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44

congestion conditions for these trucks, and truckers avoid congestion. Experts at Departments of Transportation should be consulted on the reliability of the congestion metric.

e. Terrain– State and local air agencies are required to consider terrain in the near-5 road monitoring site selection process, which in some cases may be inherently part of the roadway design. However, EPA recognizes that some states and local air agencies may have to make selections from amongst similar candidate sites that differ only by terrain, e.g. cut section versus open terrain, with or without vegetation, etc. Does the subcommittee agree that terrain and vegetation should be a consideration in the siting process? What priority should this parameter have in the overall process? NOX (and likely NO2) emissions increase with grade, but these are unlikely to be areas of heavy congestion and truck traffic. Vegetation is a sink for particles and ozone, and heavily vegetated areas should be avoided.

f. Meteorology – EPA took comment on, but did not finalize the requirement for near-road monitoring sites to be climatologically downwind of the target road segment. Reasons were because the additional limitations this would introduce in finding candidate sites would be in exchange for what may be a small increase in the opportunity to monitor peak NO2 concentrations. Further, with sites being within 50 meters of target road segments, the phenomenon of upwind meandering (pollutant transport upwind due to vehicle induced turbulence) further reduces that absolute need to be climatologically downwind. Finally, EPA recognized that, logically, the potential for peak NO2 concentration may very well occur when winds are calm or parallel (or nearly parallel) to the target road, allowing for pollutant build-up, as opposed to when winds are normal to the road. Although there is no requirement to be downwind, in the preamble to final NO2 NAAQS rule, EPA encouraged it when possible. EPA and NACAA intend to do the same in the guidance document. Does the subcommittee agree with this approach? An analysis of existing roadside CO data would be informative regarding this issue. In the peak Los Angles site, the highest levels were recorded during stagnant winter-time conditions with low-speed meandering winds, meaning that there was no consistent upwind or downwind direction.

4. Modeling is another tool that may be useful in the identification of candidate near-road sites. In particular, the use of mobile source emissions modeling with MOVES and local-scale dispersion modeling with AERMOD, can be presented as part of the guidance document. Please comment on the available modeling tools, and their pros and cons, that the subcommittee believes may be appropriate to discuss and/or recommend for use in the near-road monitoring guidance

34



1 2 3 4 5 6 7 8 9

10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43

document. Modeling requires hour-by-hour local-scale data on the fleet mix (not just HDDV counts but also the proportions of retrofitted and MY 2007-2009 and 2010+ vehicles), driving conditions (average speeds), background air quality (NO2, ozone), and meteorology (wind speed and direction, stability) that is unlikely to be available. I think a screening approach using the factors identified in my response to Charge Question #3 (with appropriate weightings developed using a literature review and analysis of existing data) should be just as useful and much easier to apply. 5. In regard to the process of identifying candidate near-road monitoring sites,

beyond the evaluation of factors noted above in question 3, and the potential use of modeling, the use of saturation monitoring and on-road monitoring are also possible tools that state and local air agencies may choose to utilize in the near-road site selection process.

a. If a state were inclined to use saturation monitoring to aid in the selection of a near-road monitoring site, and considering that the NO2 standard is a 1-hour daily maximum standard, what are the pros and cons to using passive devices to saturate an area to gather data? Given their relatively long sampling times, passive samplers would only be useful to determine likely peak 1-hour locations. In studies of pollution levels in Los Angeles communities near freeways and ports, both the passive samplers and mobile monitoring platform documented the sharp gradients in the vicinity of freeways. The Ogawa saturation monitors for NO2 had a precision of ~1.5 ppb (5%) for 1-week integrated samples, and comparison with collocated FRM analyzers indicated good accuracy (m=1.0380 and r2=0.9905).

b. Likewise, what are the pros and cons to using non-passive devices, such as near real-time or continuous devices including, but not limited to portable, non-FEM chemiluminescence methods for NO2 or Gas Sensitive Semiconductors (GSSs) for NO2 and other pollutants of interest? No comment, not my area of expertise.

c. Finally, what would be the pros and cons, to a state or local agency attempting to use a specially outfitted vehicle to collect mobile measurements to assist in the near-road site selection process for NO2 specifically as well as other pollutants of interest? The ARB has a mobile monitor that has proven to be a very useful tool to identify near-source pollutant gradients. We use an electric vehicle to avoid self pollution issues and to have a large enough battery to run all the equipment. The vehicle

35



1 2 3

5 6 7 8 9

10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44

and equipment cost are high, on the order of $250,000, and it takes expert staff to set up and operate the mobile monitor.

6. EPA recognizes that CO concentrations are primarily influenced by gasoline 4

vehicles as opposed to NO2 and PM2.5 concentrations, which are currently more heavily influenced by heavy-duty (diesel) vehicle emissions. If EPA were to propose a new set of minimum monitoring requirements for CO near roads, the near-road monitoring stations created under the implementation of the NO2 monitoring requirements may be an advantageous infrastructure for state and local air agencies to leverage. However, EPA believes there are two issues not specifically considered in the near-road NO2 monitoring language that might influence where near-road CO monitors may be most appropriately placed. The two issues are 1) the consideration of where light duty vehicles are operating under ‘cold-start’ conditions, which may often not be on the larger arterials or highways in an area, and 2) the impacts of light duty vehicle congestion and idling in areas such as urban street canyons and/or urban cores.

a. Does the subcommittee believe that the light duty cold start and congestion factors will significantly influence the location of peak CO concentrations in an area? What priority should these factors be given when compared with the factors (AADT, Fleet Mix, Roadway Design, Congestion Patterns, Terrain, and Meteorology) already being considered for peak NO2?

b. Does the subcommittee have an opinion on whether, and possibly how, these two issues of vehicles operating under cold start conditions and light duty vehicle congestion and idling in urban street canyons and/or urban cores be considered in a future, nationally applicable, CO monitoring proposal? Are there other factors that may affect peak CO concentrations and not affect peak NO2 concentrations that should also be considered for any future CO monitoring proposal? The major factors in the location of peak CO levels are high light-duty vehicle volumes, high fractions of older and poorly maintained vehicle (e.g., high emitters), stagnant meteorology, and containment by nearby buildings (e.g., urban street canyons). Parking garages can be a hot spot because of cold starts and pollutant containment by the building, but the duration of high CO periods can be short, and population exposure is generally low, so they should not be a priority for a future national CO monitoring proposal. Rather a focus on high emitters in populated areas (i.e., environmental justice communities) would be a useful complement to the consideration of factors for high NO2 levels.

7. Does the committee believe that siting considerations for identifying the location of peak NO2 concentrations will likely address all of the high priority siting considerations for PM (particularly PM2.5) as well? If not, what other factors should be considered and what are the advantages in considering these factors for identifying the location of maximum PM concentration?

36



1 2 3 4 5 6 7

9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30

The near roadway data for California from multiple studies shows that PM2.5, even in a coastal source area, is primarily from secondary formation. There was surprisingly little variation throughout the communities monitored in special studies. Several studies using ultrafine PM networks across a community showed large spatial and temporal variations associated with proximity to fresh combustion products. 8. In addition to PM2.5 mass, what other PM-related measurements are desirable at 8

near-road monitoring stations (e.g., UFP number, black carbon, EC/OC, PM coarse, etc.)?

In California special studies, we found that different BC and PM measurement methods have significant differences. They tend to agree qualitatively but not quantitatively. CPC ultrafine PM measurements are sensitive to the effective size cuts of the instruments and some instruments needed frequent calibrations. The different methods for measuring EC and OC also yield different concentrations. If near-road monitoring is going to be used for assessing compliance with air quality standards, the methods need to be comparable (equivalent) to the reference method. If near-road measurements are only to better understand the atmospheric processes and pollutant relationships, then only a good QA program is required to be useful. The most critical part of any near-road monitoring program will be the siting criteria for the monitors/samplers as a small change in distance (or possibly height), orientation to the high traffic volume direction during the light/calm winds in the morning commute, and number of vehicles during the period of typically stable air (surface inversion and light/calm wind) could all have a strong impact on the concentrations measured. 9. To allow for near-road monitoring infrastructure to be multi-pollutant, and in

reflection of the recently promulgated near-road NO2 siting criteria, reconsideration of the existing microscale CO siting criteria presented in sections 2, 6.2, and table E-4 in 40 CFR Part 58 Appendix E may be warranted. Does the subcommittee believe that reconsideration of microscale CO siting criteria is appropriate? Specifically, would an adjustment of CO siting criteria to match those of microscale PM2.5 and microscale near-road NO2 sites be logical and appropriate?

31 32 33 34 35 36 37 38 39

Yes, consistency seems warranted. 10. Even if the adjustment of microscale CO siting criteria in sections 2, 6.2, and

table E-4 in 40 CFR Part 58 Appendix E to match that of microscale PM2.5 and microscale near-road NO2 is appropriate and proposed, should there be consideration to maintain the requirement on how urban street canyon or urban core microscale CO sites should be sited?

40 41 42 43 44

37



1 2

4 5 6 7 8

10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44

No comment, not my area of expertise. 11. Does the subcommittee have an opinion on how “urban street canyons” or 3

“urban core” might be defined, perhaps quantitatively, and with regard to use in potential rule language?

No comment, not my area of expertise. 12. EPA and NACAA will select the locations for permanent sites that are part of the 9

near-road pilot study based on which state or locals volunteer to participate and can process grant funds in a timely manner to deploy equipment. From this pool of volunteers, selection should be made on certain attributes that provide the best potential to fulfill pilot study objectives. In the attached draft white paper, EPA and NACAA have proposed some potential criteria for consideration in selecting where the fixed, permanent stations should be located. These considerations include choosing a large and a relatively small urban area based on population, an area with varied or complex terrain, an urban area with an operational NOX analyzer representative of neighborhood or larger spatial scales for comparison to the near-road NOX analyzer, and an urban area with a cooperative (or non-cooperative) Department of Transportation. Does the Subcommittee agree with these considerations? Further, are there other considerations that should be evaluated in selecting pilot cities to house permanent near-road monitoring stations as part of the pilot study?

Because of the potential influence of high NO2 emissions from retrofitted or post-2007 HDDV on roadside levels, a freeway with high heavy-duty truck traffic volumes should be the primary consideration in site selection. It is also important that an ozone measurement be co-located with the neighborhood scale NOX analyzer so that total oxidant levels can be determined. USEPA, ARB, and several research groups have mobile monitors that could be deployed to quickly find the location of highest NO2. 13. EPA and NACAA have proposed that at least two urban areas should have

permanent near-road monitoring stations (that would fulfill NO2 near-road monitoring requirements) implemented for the pilot study. Please comment on the minimum equipment/pollutant measurement complement that should be deployed at each site and also the ideal equipment complement that each site should or could have, respectively. Specifically, what pollutants (e.g., NO2, NOX, NO, CO, PM (Ultrafine, 2.5, and 10), black carbon, air toxics (such as benzene, toluene, xylene, formaldehyde, acrolein, or 1,3-butadiene) and ammonia) and other information should the pilot study measure or gather at the fixed, permanent monitoring stations, and by what methods? This list should be in priority order, as feasible, and can include any NAAQS or non-NAAQS pollutant by any method (FRM/FEM and/or non-reference or equivalent methods), any particular type of other equipment for gathering supporting data such as meteorology or traffic

38



1 2 3 4 5 6

8 9

10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31

The list of pollutants is appropriate and in priority order, although air toxics could be dropped to conserve resources. Meteorological parameters (wind speed, wind direct, temperature) and light- and heavy-duty truck counts would also be useful. 14. EPA and NACAA have proposed four to five urban areas to have saturation 7

monitoring, using either passive devices and/or continuous/semi-continuous saturation type multi-pollutant monitoring packages (i.e., several types of monitors in one mountable or deployable “package”). Please provide comment on:

a. The pollutants that should be measured with the saturation devices at each saturation site.

b. The number of saturation devices per pollutant, both passive and/or continuous/semi-continuous, that may be deployed in each pilot city.

c. Whether placing saturation monitoring devices near certain road segments should include, at a minimum: 1) the highest AADT segment in an area, 2) the road segment with the highest number of heavy-duty truck/bus counts, 3) at a road segment with more unique roadway design, congestion pattern, or terrain in the area, and 4) if feasible, at a lower AADT segment with a similar fleet mix, roadway design, congestion, terrain, and meteorology as the top AADT road segment in the area. I think mobile monitoring measurements are much more useful, and more relevant to the averaging time of the NO2 standard. EPA staff can look at the details of ARB’s year-long saturation monitoring study in the communities downwind of the Ports of Los Angeles and Long Beach for information on methods, quality assurance, data analysis techniques, and results that may be useful in designing the pilot study. (http://www.arb.ca.gov/research/mobile/hcm/sat-mon/sat-mon.htm)

39



1 Dr. Kenneth Demerjian

Charge Questions 2 3

5 6 7 8 9

10 11

13 14 15 16 17 18 19 20 21 22 23

25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44

Questions regarding the near-road monitoring guidance document 15. The accompanying draft guidance document outline provides an initial thought of 4

the major topics required in the near-road monitoring guidance that will aid state monitoring agencies in the identification and implementation of NO2 near road monitoring sites from a multi-pollutant perspective. Please comment on the overall content of the recommended topics in the draft outline. Please provide suggestions on any missing subjects that should be included in the guidance document and any unnecessary topics that are currently listed in the attached draft, if applicable.

16. EPA and NACAA envision the near-road guidance document to be written from a 12 multi-pollutant perspective. What pollutants and sub-species does the subcommittee believe should be included for consideration and discussion in the near-road monitoring guidance? Some potential species for consideration include NO2, NOX, NO, CO, PM (Ultrafine, 2.5, and 10), black carbon, air toxics (e.g., benzene, toluene, xylene, formaldehyde, acrolein, or 1, 3, butadiene), and ammonia. Please prioritize the recommended pollutants and provide the rationale for their ranking, including how this pollutant measurement will contribute to scientific and regulatory knowledge of near-road air quality and adverse human health effects. Tier I - NO2, NOX, NO, CO, CO2, SO2, EC/OC, BTEX aerosol size distribution and total number concentration for routine near-road monitoring. Tier II - PM organics (HOA, OOA), NH3, HONO, H2CO and 1,3 - butadiene.

17. Identifying Candidate Near-road Site Areas 24 a. AADT & Fleet Mix – To consider fleet mix with regard to NO2, an idea is

to encourage states that have fleet mix information to take an approach that uses average, fleet-wide grams per mile emissions estimates (one for light duty vehicles and one for heavy duty vehicles), combined with AADT information to further weight which road segments in an area may be more conducive to produce peak pollutant concentrations. EPA would use the latest emission factor information to aid such a calculation. Given the variability in emission rates from on-road vehicles based on vehicle technology, fuel, speed, environmental conditions, etc., does the subcommittee believe this approach is an appropriate way to “consider” fleet mix in near-road site selection or is a more refined inventory and modeling analysis required? First cut at identifying potential near-road monitoring sites should be to consider the application of GIS methods for traffic exposure. These and other methods were recently reviewed by HEI, Traffic-Related Air Pollution: A Critical Review of the Literature on Emissions, Exposure, and Health Effects, Special Report 17, January, 2010.

b. AADT & Fleet Mix – Further, should the suggested approach above in question 4a to consider fleet mix via the use of average, fleet-wide emission factors, or the use of inventory and modeling analysis, take into

40



1 2 3 4 5 6 7 8 9

10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44

account mobile source controls that are “on the books” but have not yet been fully realized due to fleet turnover? If so, how far out into the future should states consider their effects?

c. Roadway Design – Studies suggest and support the concept that roadway design influences pollutant dispersion near the road. The EPA suggests establishing sites at-grade with the road, without any nearby obstructions to air flow; however, the Agency recognizes that this might not always be feasible. Does the subcommittee agree with this recommendation for locating sites at-grade with no obstructions? What priority should be placed on this factor within the guidance, given the need for flexibility in identifying appropriate site locations? The choice of sites with minimal roadway design influences affect pollutant dispersion should be a priority, but not an absolute requirement. Microenvironments with high pollutant exposures near local neighborhoods should be considered and will likely reflect some combination of terrain, road grade, traffic volume and congestion influences.

d. Congestion Patterns – The congestion of a roadway can be estimated by the metric “Level of Service” (LOS). LOS uses a letter grade from A to F to identify a roadway’s performance, with “A” the best conditions where traffic flows at or above the posted speed limit and all motorists have complete mobility between lanes to “F” the worst congestion where travel time cannot be predicted and generally traffic demand exceeds the facility’s capacity. Since motor vehicles generally emit more pollutants during congestion operations (although noting that NOx and select other pollutant emissions can also increase with increasing speed), how important a parameter should LOS be in the determination of appropriate near-road monitoring sites? Does the subcommittee have a view on how reliable LOS estimates are across the country? I have no firsthand knowledge on estimating LOS, but it seems that there are likely many innovative approaches that could be offered up by DOT and the traffic engineering community who deal with congestion mitigation issues on a daily basis. Discussions with this community regarding estimating LOS should be EPA’s first priority.

e. Terrain– State and local air agencies are required to consider terrain in the near-road monitoring site selection process, which in some cases may be inherently part of the roadway design. However, EPA recognizes that some states and local air agencies may have to make selections from amongst similar candidate sites that differ only by terrain, e.g. cut section versus open terrain, with or without vegetation, etc. Does the subcommittee agree that terrain and vegetation should be a consideration in the siting process? What priority should this parameter have in the overall process? Terrain and vegetation should be considered and documented as part of the siting process, but should not be a major priority factor in the selection process.

41



1 2 3 4 5 6 7 8 9

10 11 12 13 14 15 16 17 18 19 20 21

23 24 25 26 27 28 29 30 31

33 34 35 36 37 38 39 40 41 42 43 44

f. Meteorology – EPA took comment on, but did not finalize the requirement for near-road monitoring sites to be climatologically downwind of the target road segment. Reasons were because the additional limitations this would introduce in finding candidate sites would be in exchange for what may be a small increase in the opportunity to monitor peak NO2 concentrations. Further, with sites being within 50 meters of target road segments, the phenomenon of upwind meandering (pollutant transport upwind due to vehicle induced turbulence) further reduces that absolute need to be climatologically downwind. Finally, EPA recognized that, logically, the potential for peak NO2 concentration may very well occur when winds are calm or parallel (or nearly parallel) to the target road, allowing for pollutant build-up, as opposed to when winds are normal to the road. Although there is no requirement to be downwind, in the preamble to final NO2 NAAQS rule, EPA encouraged it when possible. EPA and NACAA intend to do the same in the guidance document. Does the subcommittee agree with this approach? The siting of monitors in local neighborhoods in the proximity of major roadways (i.e., <500 meters) is more important than its placement at a location that is climatologically downwind. That being said, if EPA wants to address the influence of climatology on monitor siting, the application of traditional climatological line source models would be the starting point.

18. Modeling is another tool that may be useful in the identification of candidate near-22 road sites. In particular, the use of mobile source emissions modeling with MOVES and local-scale dispersion modeling with AERMOD, can be presented as part of the guidance document. Please comment on the available modeling tools, and their pros and cons, that the subcommittee believes may be appropriate to discuss and/or recommend for use in the near-road monitoring guidance document. As mentioned above the application of traditional line source models to address siting issues with respect to climatology is fairly straight forward. The application of more sophisticated emissions and exposure models does not seem necessary to address the climatology issue.

19. In regard to the process of identifying candidate near-road monitoring sites, 32 beyond the evaluation of factors noted above in question 3, and the potential use of modeling, the use of saturation monitoring and on-road monitoring are also possible tools that state and local air agencies may choose to utilize in the near-road site selection process.

a. If a state were inclined to use saturation monitoring to aid in the selection of a near-road monitoring site, and considering that the NO2 standard is a 1-hour daily maximum standard, what are the pros and cons to using passive devices to saturate an area to gather data? Saturation monitoring for NO2 is temporally limited to 24 hr averages. The diurnal pattern of NO2 varies with season and max 1 hr averages can occur at mid-morning and mid-afternoon depending on season. Spatial mapping of NO2 (and other) concentrations with fast response monitoring technologies (e.g.

42



1 2 3 4 5 6 7 8 9

10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43

QCL multipath IR spectroscopy) can provide significant insights to near road exposures in local neighborhoods.

b. Likewise, what are the pros and cons to using non-passive devices, such as near real-time or continuous devices including, but not limited to portable, non-FEM chemiluminescence methods for NO2 or Gas Sensitive Semiconductors (GSSs) for NO2 and other pollutants of interest? QCL multipath IR spectroscopy have been demonstrated for NO2, HONO, H2CO, CO, and 1,3–butadiene and have been operated from mobile platforms providing spatial mapping or gradient measurement associated with fixed site monitoring.

c. Finally, what would be the pros and cons, to a state or local agency attempting to use a specially outfitted vehicle to collect mobile measurements to assist in the near-road site selection process for NO2 specifically as well as other pollutants of interest? This is the method of choice, but the availability of specially outfitted mobile measurement platforms is limited, as are the dollars to support such measurements.

Questions regarding the CO monitoring network and near-road monitoring 20. EPA recognizes that CO concentrations are primarily influenced by gasoline

vehicles as opposed to NO2 and PM2.5 concentrations, which are currently more heavily influenced by heavy-duty (diesel) vehicle emissions. If EPA were to propose a new set of minimum monitoring requirements for CO near roads, the near-road monitoring stations created under the implementation of the NO2 monitoring requirements may be an advantageous infrastructure for state and local air agencies to leverage. However, EPA believes there are two issues not specifically considered in the near-road NO2 monitoring language that might influence where near-road CO monitors may be most appropriately placed. The two issues are 1) the consideration of where light duty vehicles are operating under ‘cold-start’ conditions, which may often not be on the larger arterials or highways in an area, and 2) the impacts of light duty vehicle congestion and idling in areas such as urban street canyons and/or urban cores.

a. Does the subcommittee believe that the light duty cold start and congestion factors will significantly influence the location of peak CO concentrations in an area? What priority should these factors be given when compared with the factors (AADT, Fleet Mix, Roadway Design, Congestion Patterns, Terrain, and Meteorology) already being considered for peak NO2? The spatial distribution of cold start vehicles associated with urban commuting is in general broad and short term and their contribution to emissions associated with major highways adjacent to residential neighborhoods is likely small. That being said, at least one exception comes to mind. The departure of motor vehicles from major entertainment events (e.g. a football or baseball stadium) where 20-30K vehicles may be simultaneously started and caught in congestion for 10s of minutes to an hour or more. The cold start contribution, is again limited

43



1 2 3 4 5 6 7 8 9

10 11 12 13 14 15 16 17

19 20 21 22 23 24 25

27 28 29 30 31 32 33 34

36 37

in time but could contribute significantly as an emissions hot-spot impacting commuter exposes and concentrations in nearby neighborhoods.

b. Does the subcommittee have an opinion on whether, and possibly how, these two issues of vehicles operating under cold start conditions and light duty vehicle congestion and idling in urban street canyons and/or urban cores be considered in a future, nationally applicable, CO monitoring proposal? Are there other factors that may affect peak CO concentrations and not affect peak NO2 concentrations that should also be considered for any future CO monitoring proposal? The near-road NO2 concentrations are closely tied to secondary reactions with urban ozone concentrations and entrainment processes into highway line source NOx plumes. This in part, contributes to seasonal differences in near-road NO2 concentrations and its fractional contribution to NOx. Other factors affecting near-road NO2 monitoring is the distribution of gasoline and diesel vehicles. Example data analyses depicting the effects of these factors on NO2 measurements are available upon request.

Questions regarding the PM monitoring network and near-road monitoring 21. Does the committee believe that siting considerations for identifying the location 18

of peak NO2 concentrations will likely address all of the high priority siting considerations for PM (particularly PM2.5) as well? If not, what other factors should be considered and what are the advantages in considering these factors for identifying the location of maximum PM concentration? Monitoring of number concentration of ultrafine particles has spatial and temporal characteristics that do not sync all that well with that of NO2. But it remains to be seen if the health community can make the case for health outcomes for particles <100nm.

22. In addition to PM2.5 mass, what other PM-related measurements are desirable at 26 near-road monitoring stations (e.g., UFP number, black carbon, EC/OC, PM coarse, etc.)? Mobile measurement platforms are capable of performing fast response measurements (<1 minute) of key primary emission components (in addition to NO2) of interest to health effects community. These include aerosol size distribution, EC, PM organics (HOA, OOA), NH3, HONO, H2CO, CO, CO2 and 1,3 - butadiene.

Questions regarding the monitor siting criteria for microscale CO, microscale PM2.5, and the new near-road NO2 siting criteria

23. To allow for near-road monitoring infrastructure to be multi-pollutant, and in 35 reflection of the recently promulgated near-road NO2 siting criteria, reconsideration of the existing microscale CO siting criteria presented in sections 2, 6.2, and table E-4 in 40 CFR Part 58 Appendix E may be warranted. Does the subcommittee believe that reconsideration of microscale CO siting criteria is appropriate? Specifically, would an adjustment of CO siting criteria to match those of microscale PM2.5 and microscale near-road NO2 sites be logical and appropriate? The CO siting criteria should be adjusted to match those of microscale PM2.5 and near-road NO2 so there is consistency in the near-road multi-pollutant monitoring infrastructure.

38 39 40 41 42 43 44

44




1 24. Even if the adjustment of microscale CO siting criteria in sections 2, 6.2, and table

E-4 in 40 CFR Part 58 Appendix E to match that of microscale PM2.5 and microscale near-road NO2 is appropriate and proposed, should there be consideration to maintain the requirement on how urban street canyon or urban core microscale CO sites should be sited? Carry over of the urban street canyon or urban core microscale CO siting requirements should be done, keeping in mind the multi-pollutant consistency requirements mentioned in question (9).

2 3 4 5 6 7 8 9

10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44

25. Does the subcommittee have an opinion on how “urban street canyons” or “urban core” might be defined, perhaps quantitatively, and with regard to use in potential rule language? Consider emission density and spatial volume defining the urban street canyon and in the case of the urban core consider emission density and temporal persistence.

26. EPA and NACAA will select the locations for permanent sites that are part of the near-road pilot study based on which state or locals volunteer to participate and can process grant funds in a timely manner to deploy equipment. From this pool of volunteers, selection should be made on certain attributes that provide the best potential to fulfill pilot study objectives. In the attached draft white paper, EPA and NACAA have proposed some potential criteria for consideration in selecting where the fixed, permanent stations should be located. These considerations include choosing a large and a relatively small urban area based on population, an area with varied or complex terrain, an urban area with an operational NOX analyzer representative of neighborhood or larger spatial scales for comparison to the near-road NOX analyzer, and an urban area with a cooperative (or non-cooperative) Department of Transportation. Does the Subcommittee agree with these considerations? Further, are there other considerations that should be evaluated in selecting pilot cities to house permanent near-road monitoring stations as part of the pilot study? EPA should competitively fund several extramurally pilot studies in conjunction with matching state environmental monitoring funds to address this question.

27. EPA and NACAA have proposed that at least two urban areas should have permanent near-road monitoring stations (that would fulfill NO2 near-road monitoring requirements) implemented for the pilot study. Please comment on the minimum equipment/pollutant measurement complement that should be deployed at each site and also the ideal equipment complement that each site should or could have, respectively. Specifically, what pollutants (e.g., NO2, NOX, NO, CO, PM (Ultrafine, 2.5, and 10), black carbon, air toxics (such as benzene, toluene, xylene, formaldehyde, acrolein, or 1,3, butadiene) and ammonia) and other information should the pilot study measure or gather at the fixed, permanent monitoring stations, and by what methods? This list should be in priority order, as feasible, and can include any NAAQS or non-NAAQS pollutant by any method (FRM/FEM and/or non-reference or equivalent methods), any particular type of other equipment for gathering supporting data such as meteorology or traffic counts. Optimal deployment would consider one permanent near-road monitor station with a mobile measurement platform. The minimum complement of

45




1 2 3 4 5 6 7 8 9

10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32

measurement parameters at the fixed site would include NO2, NOX, NO, CO, CO2, aerosol size distribution and total number concentration, PM2.5 and PM10 mass, EC/OC, BTEX and ammonia. The minimum complement of measurement parameters for the mobile platform include aerosol size distribution, EC, PM organics (HOA, OOA), BTEX, NH3, HONO, H2CO, CO, CO2 and 1,3 - butadiene.

28. EPA and NACAA have proposed four to five urban areas to have saturation monitoring, using either passive devices and/or continuous/semi-continuous saturation type multi-pollutant monitoring packages (i.e., several types of monitors in one mountable or deployable “package”). Please provide comment on: I am not convinced this is a viable method. The only passive devices deployed systematically for saturation monitoring have been Ogawa badges for NO2. Jury is still out on saturation monitoring using battery operated PM samplers. With some R&D investment in wireless unmanned saturation sensors/samplers, routine saturation monitoring could be in the future. The proposal to use of current passive measurement devices is less than inspiring.

a. The pollutants that should be measured with the saturation devices at each

saturation site. b. The number of saturation devices per pollutant, both passive and/or

continuous/semi-continuous, that may be deployed in each pilot city. c. Whether placing saturation monitoring devices near certain road segments

should include, at a minimum: 1) the highest AADT segment in an area, 2) the road segment with the highest number of heavy-duty truck/bus counts, 3) at a road segment with more unique roadway design, congestion pattern, or terrain in the area, and 4) if feasible, at a lower AADT segment with a similar fleet mix, roadway design, congestion, terrain, and meteorology as the top AADT road segment in the area.

46



1 2 3 4 5 6 7 8 9

10 11 12 13 14 15 16 17 18

Dr. Delbert Eatough Individual response to Charge Questions for the Clean Air Scientific Advisory Committee’s (CASAC) Ambient Air Monitoring & Methods Subcommittee Peer Advisory on Near Road Monitoring To Support Measurement of Multiple NAAQS Pollutants Prepared in connection with the AAMMS committee meeting to advise EPA, September 29 - 30 in Research Triangle Park, NC. The nature of the advice which AAMMS might give to EPA on the Charge Questions outlined in the material provided to the committee will be dependent on the objectives of the multiple pollutants studies to be conducted as part of the near road monitoring program put in place in response to monitoring requirements outlined in the new NAAQS for NO2 released in February of this year (EPA CFR Parts 50 and 58, 2010). As stated by EPA in the Charge Questions document provided to AAMMS: Purpose of the Advisory 19

20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43

EPA is seeking CASAC advice on the concepts and information that should be included in the forthcoming near-road monitoring guidance document, advice on how future near-road monitoring requirements, for pollutants such as Carbon Monoxide (CO) and Particulate Matter (PM), may be drafted in a way to mesh with the existing Nitrogen Dioxide (NO2) requirements and foster a multi-pollutant monitoring infrastructure, and the objectives, approach, and execution of the near-road monitoring pilot study. This is a bold new direction being taken by EPA which will move the concept of multi-pollutant monitoring in support of Clean Air objectives forward in a significant way. While reasonable detail is given in the charge questions related to the approach to be used and execution of the near-road monitoring pilot study, little detail is given on the scientific objectives of both the pilot study and the near-road monitoring program which is required under the NO2 NAAQS. I will start these comments by framing some of my thoughts on the possible objectives of this program, with the belief that the nature of the advice which might be given in very dependent on the identified objectives. I should emphasis that the literature cited in my comments is illustrative only and not intended to be a complete review of what is currently know. I. Objectives of the Near-Road Monitoring Program. A. NO2 Monitoring Time Scale.

47



1 2 3 4 5 6 7 8 9

10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43

As outlined by EPA, the advice must consider the near-road monitoring requirements of the NAAQS for NO2, which is to have ambient monitoring conducted at the location of maximum NO2 concentrations in an area, which at a minimum is directly attributable to mobile source emissions. While not explicitly stated in the charge questions, I assume that this means that the monitoring to be conducted will be focused on the 1-hour time period requirement of the NAAQS. This is an important point because the information one can gain from the near-road monitoring program is dependent on the time period chosen for study. For example, 1-hour average monitoring allows the identification of the effects of many diurnal variations which the identification of such details as diurnal changes in sources and atmospheric processes (Eatough, 2008). However, there are important processes which occur on a much shorter time period which will not be as well identified in the program (Zhu 2002a, b). My comments here are limited to considerations for a 1-hour average monitoring program. B. NO2 Monitoring Objectives Other than Identification of Maximum Concentrations. If we were discussing the identification of the maximum concentration of TSP Pb, the sampling objectives would be relatively straightforward because particulate Pb is a species which is expected to be relatively conserved in the atmosphere after emission. However, the story is much more complex with respect to NO2 if one wants to know, in addition to identification of maximum concentrations, the atmospheric factors which contributed to the maximum concentrations observed and the effect of the NO2 formation pathways on other NAAQS pollutants such a ozone, or the effect of ambient ozone on the formation of NO2. The great majority of primary nitrogen oxides are currently emitted from combustion sources in the form of NO (g) (Finlayson-Pitts, 2000). In the presence of ozone, HO2 or RO2 , NO is oxidized to NO2, NO (g) + O3 (g → NO2 (g) + O2 (g) (1) )

NO (g) + RO2 (g) → NO2 (g) + RO (g) (3)

NO (g) + HO2 g) → NO2 (g) + OH (g) (2)

48

Ozone, in turn is formed from the photolysis of NO2, OH from the photolysis of O3 and HO2 and RO2 from the reaction of OH (and at night NO3 ) radicals with gas phase organic compounds. Thus, the relative concentrations of NOX and gas phase organic compounds control the concentration of ozone in a complex manner described by ozone isopleths (Finlayson-Pitts 2000), and, in turn, these processes control the concentrations of NO2. In general, at low NOX concentrations the concentration of ozone is little effected by the concentration of VOC and the system is NOX limited. However, at low concentrations of VOC, the concentrations of ozone can decrease with increasing NOX concentrations as NO reacts with ozone and NO2 competes with VOC for the OH radical



1 2 3 4 5 6 7 8 9

10 11 12 13 14 15

22

25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

by the irreversible formation of nitric acid, NO2 (g) + OH (g) → HNO3 (g) (4) Concentrations of NO2 observed at a site will be effected by this complex chemistry. Complete understanding of the etiology of NO2 concentrations identified in a near-road monitoring program will require the identification of each of these factors (Kuprov 2010). I have assumed that understanding these chemical contributions to the observed NO2 concentrations will be one of the monitoring objectives. C. Multi-Pollutant Monitoring Objectives. The multi-pollutant monitoring portion of the near-road monitoring program plan being developed for the NO2 requirements could have several objectives: $ Identification of concentrations of other NAAQS pollutants which accompany the 16

observed NO2 concentrations. 17 $ Identification of the contribution these and other key pollutants make to the 18

observed NO2 concentrations (see my comments in B.) 19 $ Identification of the atmospheric processes which contribute to the observed 20

concentrations of NO2 and the other monitored pollutants (e.g. Kuprov, 2010; 21 Wilson 1977).

$ Identification of the sources which contribute to both measured concentrations of 23 NO2 and the other measured NAAQS pollutants (e.g. Eatough 2008). 24

My comments assume that meeting all of these objectives is important in the design of the program. Another possible objective of the monitoring program is the identification of toxic compounds to inform health studies. While this is an important objective, it is somewhat less directly related to the above and I have not given this objective high priority. I have not attempted to frame an individual response here to all charge questions as there are some where I am not an expert. The charge questions are given in the consensus report. II. Response to the Charge Questions. Charge Question 1. I suggest the following are areas where the Guidance Document may need strengthening above what I think is intended in the outline: $ Introduction: Based on the material in the Pilot Study draft, it is not clear that 43

EPA has yet identified the reasons (scientific objectives) for the multi-pollutant 44

49



1 2 3 4 5 6

9

12 13 14 15 16

19

22 23 24 25 26 27 28 29

31

34 35 36 37 38 39 40 41 42 43 44

$ Background. The background should also contain the scientific basis for the 7

multi-pollutant objectives, as outlined in the bullet above. 8 $ Identifying Candidate Near-road Site Areas. I believe the criteria outlined for this 10

section are adequate for the identification of a site where maximum NO2 11 concentrations near a given near-road site may be determined. However, since NO2 is a secondary pollutant and its concentrations will be effected by both emissions from the roadway and from any other nearby elevated sources (VOC, ozone, etc.). These factors should be considered in the site identification process.

$ Modeling. I am not an expert in this area. However, modeling should take into 17

account the factors I have discussed above. 18 $ Monitoring. One potential problem with saturation sampling is that if all data are 20

not collected under identical conditions, certainly with respect to time and traffic 21 flow, the comparison of results for the various saturation samplers may not be meaningful. I am also concerned about this issue in connection with the use of mobile monitoring. How will assurance be obtained that a comparison of measurements at two locations at different times gives the same result as a comparison of measurements at two different sites at the same time. How will diurnal variability be taken into account? These issues are discussed in the consensus report.

$ Near-road Site Selection. The items outlined here seem reasonable. 30 $ Recommended Near-road Site Documentation. The adequacy of this section will 32

be dependent on the adequacy of the EPA objectives for the Pilot Study in 33 addressing the appropriate multi-pollutant monitoring objectives. This will be an area discussed under other Charge Questions.

Charge Question 2. I have suggested in I.C. objectives that should be part of the multi-pollutant monitoring scheme. My thoughts on species which should be included to meet each of these scientific objectives (the objectives are repeated here) are given below. Again, all these measurements need to be made on a one-hour time basis. It is also recognized that most of these measurements will not be made at all sites but only at the limited number of advanced sites as discussed in the consensus document.

50



1

4 5

8 9

10

13 14

17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33

38 39 40 41 42 43 44

$ Identification of concentrations of other NAAQS pollutants which accompany the 2

observed NO2 concentrations. (CO, PM10, PM2.5, Ozone and {probably for some, 3 but not all sites} sulfur dioxide)

$ Identification of the contribution these and other key pollutants make to the 6

observed NO2 concentrations (see my comments in B.) (VOC related to ozone 7 formation, NOX, NOY {including a minimum of gas and particulate nitrate in addition to NOX}).

$ Identification of the atmospheric processes which contribute to the observed 11

concentrations of NO2 and the other monitored pollutants (e.g. Kuprov, 2010, 12 Wilson 1977). (The species listed in the two proceeding bullets.)

$ Identification of the sources which contribute to both measured concentrations of 15

NO2 and the other measured NAAQS pollutants (e.g. Eatough 2008). (Fine 16 particulate OC and EC, BC and UV C. In addition techniques are now becoming available for the hourly measurement of fine particulate elements and organic markers on an hourly basis. These last two measurements would be lower priority, but where they can be measured would greatly add to meeting this objective).

I have not listed any of the toxic gases included in the charge question, but they are relevant to health objectives and might be added if EPA wants to add an objective for this specific purpose. That is a little different than the atmospheric chemistry objectives on which I have focused. Charge Question 5. Saturation monitoring can aid greatly in the identification of a suitable near-road monitoring site. My only concern with respect to this Charge Question is that it be made clear that the saturation monitoring must meet two key objectives: $ The data must be available on a one-hour average basis, consistent with NAQQS 34

requirement for NO2. 35 $ All saturation data must be obtained at all locations on the same time basis so the 36

results are not significantly confounded by the diurnal and seasonal variations in 37 NO2 emissions and formation chemistry.

Because of the inherent problems in meeting the items in the above two bullets, care needs to be taken in the use of an outfitted vehicle to assist in the road-site selection process. Charge Question 7.

51



1 2 3 4 5 6 7 8 9

10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25

31 32 33

40 41 42 43 44

While peak concentrations of ultrafine particles will frequently be associated with emissions from vehicles, the concentrations of PM2.5 will not. In almost all urban studies I am aware of, the maximum concentration of PM2.5 are not dominated by primary emissions but the secondary formation of nitrate and organic material, and in the east by regional sulfate. None of these contributions can be elucidated from near-road monitoring. However, the total pollutants suggested here to be monitored in the program can inform the secondary formation processes which lead to these elevated PM concentrations. Charge Question 8. I have outlined my thoughts on this charge question in the response to Charge Question 2. Charge Question 13. I have listed my thoughts on equipment needed as outlined in this Charge question in my response to Charge Question 2, with an indication of contributions to be expected for each measurement. The relative priority which might be assigned depends on whether or not EPA agrees with my outline of objectives in Section I. It is recognized that these measurements will be made at a limited number of sites. My priority order and suggested measurements are: 1. Top priority (measurement of NO2 and NAAQS pollutants): $ NO2, hourly averaged data by an artifact free measurement as well as an FRM or 26

FEM technique. 27 $ Ozone, hourly averaged data by an FRM or FEM technique. 28 $ PM10 and PM2.5 by a dichot FDMS TEOM method (to avoid the loss of volatile 29

material). 30 2. Second priority (measurement of species which will inform NO2 chemistry. $ VOC, hourly averaged data. 34 $ NOX, hourly averaged data by an FRM or FEM technique. 35 $ Ozone hourly averaged data by an FRM or FEM technique (also listed in 1.). 36 $ NOY hourly averaged data. 37 $ Nitric acid and particulate phase nitrate, hourly averaged by an IC technique (e.g., 38

the URG AIM) where nitrate is known to be high, e.g. LA or western mountain 39 valleys.

3. Third priority (data to aid in source apportionment, including separation of

gasoline and diesel vehicle contributions).

52



5 6 7 8 9

10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44

$ Hourly average EC and OC, preferable by a Sunset dual oven instrument. 1 $ BC and UV hourly average Aethalometer data. 2 $ Hourly averaged fine particulate elemental and trace organic marker data. 3

4 Charge Question 14. First a general comment. The saturation studies are intended to aid in the identification of near-road sites which will give maximum NO2 concentrations. These then will become the site(s) which are used to meet the NO2 near-road monitoring requirements. To meet this requirement the key data each saturation study must provide are hourly average NO2 concentrations which define at least a couple of weeks diurnal variation in the NO2 concentrations. Less than hourly and less than complete diurnal coverage will not truly inform on maximum concentrations. I am not certain that a passive device can meet this need, so I assume a semi-continuous device or a modified passive sampler (see the consensus report) would be used. a. NO2

. b. I would think 4 to 6 is a reasonable number, but I defer to others who have conducted saturation studies. c. Of the criteria listed in the charge question, 1) and 2) seem most important. I would also pick a site where impact from VOCs nearby is important. REFERENCES Eatough, D.J., Grover, B.D., Woolwine, W.R., Eatough, N.L., Long, R., and Farber, R. (2008). “Source Apportionment of 1-hr Semi-Continuous Data from Riverside During SOAR 2005 Using Positive Matrix Factorization.” Atmos. Environ. 42: 2706-2719. EPA (2010). 40 CFR Parts 50 and 58, Primary National Ambient Air Quality Standards for Nitrogen Dioxide; Final Rule, Federal Register, February 9, 2010, pp 6474-6537. Finlayson-Pitts B.J. and Pitts, J.N. Jr. (2000) Chemistry of the Upper and Lower Atmosphere: Theory, Experiments and Applications, Academic Press. Kuprov R., Eatough D.J., Hansen J.C., Cruickshank T. And Olson N. (2010) Composition and Secondary Formation of Fine particulate Material in the Salt Lake Valley: Winter 2009, J. Air & Waste manage. Assoc., in press.

53



1 2 3 4 5 6 7 8 9

10 11 12 13

Wilson, W.E., Spiller, L.L., Ellestad, T.G., Lamothe, P.J., Dzubay, T.D., Stevens, R.K., Macias, E.S., Fletcher, R.A., Husar, J.D., Husar, R.B., Whitbe, K.T., Kittelson, D.B. and Cantrell, B.K. (1977) “General Motors sulfate Dispersion Experiment: Summary of EPA Measurements.” J. Air Pollut. Cont. Assoc. 27: 46-51. Zhu, Y., Hinds, W.C., Kim, S., and Sioutas C. (2002a) “Concentration and Size Distribution of Ultrafine Particles Near a Major Highway.” J. Air & Waste Manage. Assoc. 52: 1032-1042. Zhu, Y., Hinds, W.C., Kim, S., Shen S., and Sioutas C. (2002b) “Study of Ultrafine Particles Near a Major Highway with Heavy-duty Diesel Traffic.” At. Environ. 36: 4323-4335.

54



1 2 3

5 6 7 8 9

10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44

Mr. Dirk Felton

Questions regarding the near-road monitoring guidance document 1. Please comment on the overall content of the recommended topics in the 4

draft outline. Please provide suggestions on any missing subjects that should be included in the guidance document and any unnecessary topics that are currently listed in the attached draft, if applicable.

The background section should include language suggesting that finding one near-road location where all pollutants of interest have the highest concentrations is not likely and compromises will be necessary and acceptable. For example, NO2 concentrations may well be higher in dense urban neighborhoods away from well ventilated busy highways. This section should also include a discussion of the limitations of the various monitoring methods in the near-road environment. The PM-2.5 FRM has demonstrated poor capture efficiency for volatile fresh emissions from mobile sources and CO monitoring for a health based NAAQS near roadways may not be warranted. Modeling and saturation or mobile monitoring should not be required in the site selection process. This type of work is beyond the capacity of many monitoring agencies, and since these methods have not been uniformly demonstrated or well documented, they are not likely to provide much assistance. If monitoring agencies have existing information that could provide this type of information, they certainly should consider the information and make it available to their Regional EPA office during the site approval process. The last section on site documentation should not be burdensome to the monitoring agencies. Much of the “NCore type” of site characterization documentation was designed so that EPA staff would have a convenient way to review monitor siting. If this type of information is important to the EPA then they can collect the data when they visit the sites.

2. EPA and NACAA envision the near-road guidance document to be written from a multi-pollutant perspective. What pollutants and sub-species does the subcommittee believe should be included for consideration and discussion in the near-road monitoring guidance? Some potential species for consideration include NO2, NOX, NO, CO, PM (Ultrafine, 2.5, and 10), black carbon, air toxics (e.g., benzene, toluene, xylene, formaldehyde, acrolein, or 1, 3, butadiene), and ammonia. Please prioritize the recommended pollutants and provide the rationale for their ranking, including how this pollutant measurement will contribute to scientific and regulatory knowledge of near-road air quality and adverse human health effects.

The discussion of potential species for consideration in the near road monitoring guidance is complicated by the currently available methods for some of these pollutants. PM and specifically PM-2.5 would certainly be considered to be a candidate but the EPA does not have an acceptable method for measuring PM-2.5 in the near-road environment. The EPA should consider the development of a suitable method for PM-2.5 to be of the utmost necessity. CO monitoring is recommended and the siting requirements should be modified to be identical to the near road NO2 requirements. CO may or may not be found at high

55



1 2 3 4 5 6 7 8 9

10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44

enough concentrations to be a health issue, but it will be useful as a tracer of primary emissions. Ultrafine monitoring is recommended but an appropriate minimum and maximum size range must be considered. Ultrafine particles have been found to affect health and data from near road sites will assist in the understanding of how mobile sources interact with ambient ultrafine levels. BC is a combustion tracer so it should only be contemplated for installation at a near road site if there is another BC instrument within the same CBSA. Since the NATTs program requires BC, it makes sense to only install these instruments in the near road sites that also have a NATTs site. This limited deployment should provide some information on the gradient of BC between the near road environment and a central monitor. This program should include air toxics that have been identified as known human carcinogens. This would include benzene, formaldehyde and 1, 3 - butadiene. Air toxics monitoring at the near road air sites should be encouraged in the CBSAs that also have a NATTs site. This would allow for comparisons between the near road and community scale monitors. Over time, the changes in vehicle technology and fuels will continue to impact concentrations of these air toxics which large segments of the population are exposed to daily. Currently, concentrations of these specific air toxics already exceed acceptable concentrations in many areas of the country. A critical evaluation of the impacts of these changes on the concentration of these air toxics in urban areas across the U.S. is extremely important.

3. Identifying Candidate Near-road Site Areas a. AADT & Fleet Mix – does the subcommittee believe this approach is

an appropriate way to “consider” fleet mix in near-road site selection or is a more refined inventory and modeling analysis required?

Monitoring agencies should be permitted to use the best available information in order to help with site selection. There are too many disadvantages to specifying a one size fits all standard approach to site selection. Each CBSA has its own set of variables such as roadway restrictions on vehicle type or vehicle type at certain hours of the day, tolls that vary as a function of vehicle type, bridge restrictions, weather conditions such as cold weather or high winds and differences in required vehicle emission controls. Additionally, congestion in dense urban areas with numerous roadways in close proximity to one another can lead to higher NO2 concentrations than AADT and fleet mix would indicate. Some monitoring agencies also have very good inventories and existing modeling work done for specific CBSAs and not as much information for others. The guidance document should encourage the monitoring agencies to utilize all available tools and sources of information to help select the most appropriate monitoring locations.

b. AADT & Fleet Mix – Further, should the suggested approach above in question 4a to consider fleet mix via the use of average, fleet-wide emission factors, or the use of inventory and modeling analysis, take into account mobile source controls that are “on the books” but have not yet been fully realized due to fleet turnover? If so, how far out into the future should states consider their effects?

56



1 2 3 4 5 6 7 8 9

10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44

Ambient monitoring is concerned with the current exposure to the population from sources of pollutants. Fleet turnover in most cases is a long term process especially for the largest trucks unless there is a specific program in place to help replace these vehicles more quickly. The EPA has the ability to review monitor siting once a year when the monitoring agencies submit their annual network plan. This should provide ample opportunity for the EPA to suggest changes to the monitoring network in light of changes in AADT and fleet turnover.

c. Roadway Design – Does the subcommittee agree with this recommendation for locating sites at-grade with no obstructions? What priority should be placed on this factor within the guidance, given the need for flexibility in identifying appropriate site locations?

The design of the roadway is extremely significant and it makes sense to recommend that the preferred installation be at grade with no obstructions. That said, the guidance document should expand on why this is important and where low or high concentrations are likely to be found. Low concentrations may be found adjacent to roadways elevated on piers, near coastlines or large water bodies. High concentrations may be found adjacent to roadways situated in dense urban areas surrounded by tall buildings, next to below grade roadways or near roadways that are also influenced by tunnel ventilation systems. These high concentration locations may be the preferred locations for some CBSAs particularly if they are also significant for population exposure.

d. Congestion Patterns – how important a parameter should LOS be in the determination of appropriate near-road monitoring sites? Does the subcommittee have a view on how reliable LOS estimates are across the country?

It is likely that on stagnant days the roadways with poor level of service, including congested slow moving traffic will cause levels of pollutants to accumulate to higher than expected levels. LOS information should be used to qualify the AADT data on a CBSA specific basis.

e. Terrain – Does the subcommittee agree that terrain and vegetation should be a consideration in the siting process? What priority should this parameter have in the overall process?

Terrain and vegetation are very important and should rank near the top of the site selection criteria. Both of these factors drastically affect path length, population exposure, effective probe height and an agency’s ability to site a monitor. Vegetation can be a sink of many pollutants as well as a screen for the efficient transport of pollutants between the source and the monitor.

f. Meteorology - Although there is no requirement to be downwind, in the preamble to final NO2 NAAQS rule, EPA encouraged it when possible. EPA and NACAA intend to do the same in the guidance document. Does the subcommittee agree with this approach?

The importance of other factors such as AADT, terrain and population exposure should take priority over the prevailing wind direction. If the monitoring agency has candidate locations that are otherwise equal it is preferred to select the one that is predominantly downwind from the near road emissions.

57



2 3 4 5 6 7 8

10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43

4. Modeling is another tool that may be useful in the identification of candidate 1 near-road sites. In particular, the use of mobile source emissions modeling with MOVES and local-scale dispersion modeling with AERMOD, can be presented as part of the guidance document. Please comment on the available modeling tools, and their pros and cons, that the subcommittee believes may be appropriate to discuss and/or recommend for use in the near-road monitoring guidance document.

5. In regard to the process of identifying candidate near-road monitoring sites, 9

beyond the evaluation of factors noted above in question 3, and the potential use of modeling, the use of saturation monitoring and on-road monitoring are also possible tools that state and local air agencies may choose to utilize in the near-road site selection process.

General Comment: All of these tools are resource intensive and are likely to be beyond the scope of many monitoring agencies. The use of these tools should not be required.

a. what are the pros and cons to using passive devices to saturate an area to gather data?

The NO2 standard is a 3-Yr average of 1-Hr maximum values, so a typical passive sampler which is exposed for days to weeks will provide better information if the traffic signature is relatively consistent throughout the course of a day. This is likely to be the case in the largest CBSAs. Passive samplers will not provide as much information in smaller CBSAs with variable traffic congestion and the affects of other significant sources of pollutants.

b. Likewise, what are the pros and cons to using non-passive devices, such as near real-time or continuous devices including, but not limited to portable, non-FEM chemiluminescence methods for NO2 or Gas Sensitive Semiconductors (GSSs) for NO2 and other pollutants of interest?

The concept is good but, of course, the real issue with many of these instruments is the comparability to data from the FRM. If a monitoring agency is able to use several of these instruments to look at potential sites in relation to each other, not to a nearby FRM, then these non-regulatory instruments could be useful.

c. Finally, what would be the pros and cons, to a state or local agency attempting to use a specially outfitted vehicle to collect mobile measurements to assist in the near-road site selection process for NO2 specifically as well as other pollutants of interest?

This could be a very expensive component of the site selection process and would only be likely to occur if the monitoring agency already had access to this type of vehicle. The advantage is that candidate areas can be identified relatively quickly compared to other screening methods. The disadvantages to these types of measurements are that they are a snapshot in time that could have little relationship to the periods when the concentrations are expected to be the highest. Questions regarding the CO monitoring network and near-road monitoring

58



2 3 4 5 6

8 9

10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44

6. The two issues are 1) the consideration of where light duty vehicles are 1 operating under ‘cold-start’ conditions, which may often not be on the larger arterials or highways in an area, and 2) the impacts of light duty vehicle congestion and idling in areas such as urban street canyons and/or urban cores. Questions regarding the PM monitoring network and near-road monitoring

7. Does the committee believe that siting considerations for identifying the 7 location of peak NO2 concentrations will likely address all of the high priority siting considerations for PM (particularly PM2.5) as well? If not, what other factors should be considered and what are the advantages in considering these factors for identifying the location of maximum PM concentration?

The answer to this question is complicated by the fact that the NO2 regulation incorrectly identifies near-road environments as the areas where NO2 concentrations will always be highest. Secondly, the traditional method of measuring PM-2.5, the FRM is not well suited to capturing the highly volatile emissions emitted from mobile sources. Thirdly, a significant portion of PM-2.5 in most areas is either due to transport or due to secondary particle formation, neither of which have anything to do with micro-scale NO2 siting criteria. A PM-2.5 network design that only includes locations that are near-road micro-scale sites will only provide information about that source. An adequate PM-2.5 network must include sites that also provide information that are relevant for upwind assessment, population exposure, seasonal differences that can be evaluated in terms of particle formation and lastly micro-scale source attribution. PM-2.5 is more likely to be highest in dense urban neighborhoods away from the well ventilated roadways. These areas are subject to emissions from transport, domestic heating and cooking, stationary sources and mobile sources.

8. In addition to PM2.5 mass, what other PM-related measurements are desirable at near-road monitoring stations (e.g., UFP number, black carbon, EC/OC, PM coarse, etc.)?

The EPA must define how it intends to collect PM-2.5 mass. The FRM will miss a significant fraction of the volatile component of the mobile source emissions. This will create the situation where the apparent risk from these sources or from living in one of these areas is under estimated. In general, it is not advisable to err on the side of underestimating a health risk when establishing a NAAQS oriented monitoring program. It is preferable to delay a requirement to monitor PM-2.5 at these locations until the EPA develops a suitable method that reliably includes a majority of the volatile component of PM-2.5. One suggestion is for the EPA to encourage the manufacturers of continuous PM-2.5 monitors to develop instruments without the FEM algorithms that reduce PM concentrations in an attempt to emulate the filter based FRM. This will provide a more realistic indication of the actual PM concentrations that the populations in these areas are exposed to everyday. UFP number monitoring is on its way to becoming an acceptable monitoring technique but it is not quite ready for routine use. The EPA should invest in limited deployment

59



1 2 3 4 5 6 7 8 9

10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44

monitoring demonstrations to assist the vendors as they develop better more reliable instrumentation. Some of the issues that need to be resolved include, how small do we need to go, which bins are appropriate from a health perspective and how do we QA these instruments. UFP data could provide valuable information for the health community as they investigate air quality related health effects. The EPA has a BC monitoring requirement in place at the NATTs sites but this data is rarely used. The data is also subject to artifacts due to filter changes and interference from co-pollutants on the filter substrate. The EPA should really determine if there is a need for this data before the requirement to monitor for a non-criteria pollutant is added to this network. EC/OC data is more interesting because it provides information that can help regulators understand the losses from the PM-2.5 FRM and to develop potential pollution control programs. This information may only be necessary in the MSAs where the PM NAAQS could be exceeded and control strategies are required. PMcoarse is of limited value because the quality of the PM-2.5 data is poor and because of the small scale that the resultant data represents. Questions regarding the monitor siting criteria for microscale CO, microscale PM2.5, and the new near-road NO2 siting criteria

9. Does the subcommittee believe that reconsideration of microscale CO siting criteria is appropriate? Specifically, would an adjustment of CO siting criteria to match those of microscale PM2.5 and microscale near-road NO2 sites be logical and appropriate?

The 3 questions (9,10 and 11) assume that there are high enough CO concentrations at typical near-road locations to justify including health related NAAQS CO monitors at these sites. The EPA must determine if this is the case before establishing a new, expensive and potentially un-necessary monitoring requirement. The siting criteria for CO does need to be updated if CO is included at a near-road monitoring location. The more restrictive height requirement is not necessary for sites where turbulence is expected to create well mixed conditions at these monitors. The criteria used for micro-scale PM and NO2 is acceptable.

10. should there be consideration to maintain the requirement on how urban street canyon or urban core microscale CO sites should be sited?

The micro-scale PM and NO2 siting requirements are adequate for street canyon CO monitoring. The existing CO canyon monitoring guidance included a wind direction provision that is un-necessary and made it more difficult to find suitable locations.


It is useful to be able to refer to an area as a street canyon or as an urban core. Street canyons are relatively easy to define because they are defined by the structures in their immediate vicinity. Urban Cores have to be defined on a larger MSA basis. Street Canyons have the following attributes. They are relatively narrow in comparison to the height of the structures on either side of the street. It should be relatively easy to

60



1 2 3 4 5 6 7 8 9

10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43

model an optimum or minimum width to height ratio that would define a street canyon. A ratio of (1) width to (1 or 1.5) height might be a good starting point. Urban Cores can be defined by a number of parameters such as the reasonable geographic center of an MSA, the approximate centroid of emissions sources, area of densest population or highest congestion. Very large MSAs could easily have multiple urban cores due to geography and neighborhood layout such as in NYC where boroughs in some ways can emulate small cities. Questions regarding the near-road monitoring pilot study The EPA ORD has been operating near-road monitoring locations for more than a year in a couple of locations including Las Vegas. The results from these campaigns should be used to determine if the NO2 and other pollutant concentrations were high enough to warrant further development of near-road monitoring efforts. The data should also be evaluated to determine if the selected monitoring methods were appropriate for these locations.

12. Does the Subcommittee agree with these considerations? Further, are there other considerations that should be evaluated in selecting pilot cities to house permanent near-road monitoring stations as part of the pilot study?

The reliance on a CBSA’s population is too simplistic. The population of a CBSA does not specifically provide information about how roadways are used within a CBSA. For example, this ranking does not include the number of people who commute from outside of the CBSA every day or use car pools or mass transit including busses, trains, subways and ferries.

13. Please comment on the minimum equipment/pollutant measurement complement that should be deployed at each site and also the ideal equipment complement that each site should or could have, respectively. Specifically, what pollutants (e.g., NO2, NOX, NO, CO, PM (Ultrafine, 2.5, and 10), black carbon, air toxics (such as benzene, toluene, xylene, formaldehyde, acrolein, or 1,3, butadiene) and ammonia) and other information should the pilot study measure or gather at the fixed, permanent monitoring stations, and by what methods? This list should be in priority order, as feasible, and can include any NAAQS or non-NAAQS pollutant by any method (FRM/FEM and/or non-reference or equivalent methods), any particular type of other equipment for gathering supporting data such as meteorology or traffic counts.

It is acceptable to deploy more monitoring parameters at the pilot sites than at the routine sites, however, the EPA must be aware of the resources necessary to operate these sites. The pilot sites should have NO2, CO, ultrafine and BC monitors. PM-2.5 should be included only if PM-2.5 FEMs are operated without the algorithms that reduce the mass to match the FRM. These instruments could utilize the EPA parameter description 88500 which cannot be compared to the PM2.5 NAAQS and is defined as total atmospheric PM-2.5. The pilot sites should be selected based on the availability of NCore and NATTs sites within the same CBSA to permit data comparisons with the near-road site.

61



2 3 4 5 6 7 8 9

10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

14. EPA and NACAA have proposed four to five urban areas to have saturation 1 monitoring, using either passive devices and/or continuous/semi-continuous saturation type multi-pollutant monitoring packages (i.e., several types of monitors in one mountable or deployable “package”). Please provide comment on:

The deployable semi-continuous package that EPA is considering is not practical for use by monitoring agencies. The unit is expensive, may or may not provide data that is comparable to criteria monitors and would require power and security wherever it is deployed.

a. The pollutants that should be measured with the saturation devices at

each saturation site. The saturation devices only have to be able to monitor NO2. Once the high NO2 site is found, other parameters can be included when the monitoring site is fully established. b. The number of saturation devices per pollutant, both passive and/or

continuous/semi-continuous, that may be deployed in each pilot city. The most effective use of saturation samplers is to help rank the top 2-4 sites that have been identified through the rest of the site selection process. It would be too expensive and labor intensive to use saturation monitoring to select sites across wide areas in a CBSA. c. Whether placing saturation monitoring devices near certain road

segments should include, at a minimum: 1) the highest AADT segment in an area, 2) the road segment with the highest number of heavy-duty truck/bus counts, 3) at a road segment with more unique roadway design, congestion pattern, or terrain in the area, and 4) if feasible, at a lower AADT segment with a similar fleet mix, roadway design, congestion, terrain, and meteorology as the top AADT road segment in the area.

It is sensible to use saturation monitors to compare sites that cannot be ranked in another way. The monitoring agency should include the feasibility of establishing a monitor in a specific area in the initial consideration of potential sites. Saturation samplers can help the agency differentiate a feasible site on one type of roadway to another feasible site on another type of roadway.

62



1 2 3 4 5 6 7 8 9

10 11 12 13 14 15 16 17 18 19 20

Dr. Kazuhiko Ito Note: The charge questions below are truncated to save space. General Comment: I understand the importance of measuring NO2 near roadways, but I am very concerned about a possible reduction in the number of community-wide monitors that can result from the new network plan. According to EPA, there are currently about 400 NO2 monitors nationwide, many of which are community-wide monitors, and very few (3?) are micro-scale monitors. The new network plan will create 126 near-road monitors, but only 53 community-wide monitors will be required to operate. The epidemiological studies that reported short-term associations between NO2 and respiratory morbidity used the community-wide monitors. Because the near-road monitors may not adequately reflect the entire city’s residents’ exposures, we may be reducing the number of NO2 monitors that can be used for community based epidemiological studies. I have more comments on this as part of response to Charge Question 2. Questions regarding the near-road monitoring guidance document Charge Question 1: Please comment on the overall content of the recommended topics in the draft outline (of the near-road monitoring guidance document). Please provide suggestions on any missing subjects that should be included in the guidance document and any unnecessary topics that are currently listed in the attached draft, if applicable.

21 22 23 24 25

Comment: 26 27 28 29 30 31 32 33 34 35 36 37 38 39

The overall content and the topics listed in the outline look generally adequate. I

have a few comments below. In addition to the literature review mentioned as part of Background, EPA can

conduct analysis of the existing data (though the monitors are not sited in the required near-road scale) to describe the relationship between NO2 and other pollutants (e.g., CO, EC, etc.).

In identifying candidate near-road areas, it may not be just AADT but the density

of high AADT areas that are important, particularly in cities with high population density areas. Charge Question 2: What pollutants and sub-species does the subcommittee believe should be included for consideration and discussion in the near-road monitoring guidance? Please prioritize the recommended pollutants and provide the rationale for

40 41 42

63



1 2

their ranking, including how this pollutant measurement will contribute to scientific and regulatory knowledge of near-road air quality and adverse human health effects.

Comment: 3

4 5 6 7 8 9

10 11 12

13 14 15 16 17 18 19 20

21 22 23 24 25

26 27 28

29 30 31 32 33 34

I need to first comment on my general rationale of measuring co-pollutants at the proposed near-road monitoring location before discussing the prioritizing co-pollutants. I feel that there is a gap between the recognition of the short-term associations between NO2 and respiratory morbidity in the observational epidemiological studies, the results from human laboratory studies, and the suggestion to measure hourly NO2 values at near-road locations and set a standard level at such locations with the 1-hr averaging time. I understand that this has been already decided, but I think we should keep the options to evaluate the implication of this decision in case we may need to reconsider this strategy. The issues I need to raise are the following:

• The observational studies that reported short-term associations between NO2 and respiratory morbidity mentioned in the FR (40 CFR Parts 50 and 58) used NO2 data from monitors that were not near-road monitors. Unless the associations were due to exacerbations of respiratory conditions in the sub-populations who lived near roads, it is possible that the data to be collected at near-road monitors may not correlate well with the measurement at the existing NO2 monitors, and they may not even correlate well with the citywide respiratory morbidity time-series because the near-road measurements may be highly influence by local sources that are not relevant to the rest of the city.

• Correlations between NO2 and co-pollutants at near-road monitors may be different from those at non-near-road monitors (i.e., those used in the observational epidemiological studies). Therefore, it is possible that these co-pollutants measured at near-road monitors have different impacts on the health effects models compared to those reported in the past epidemiological studies.

Because of these issues, I think it is essential to retain at least some of the existing non-NR NO2 monitors until the issues are resolved. Otherwise, it is possible that we end up with discontinuation of data that are useful for epidemiological studies.

The review of epidemiological studies in the NO2 ISA concluded that the associations between NO2 and respiratory morbidity were robust to the inclusion of other co-pollutants (i.e., no strong indication of confounding) in the health effects models, but these co-pollutants were mostly other criteria pollutants (PM10, PM2.5, ozone, CO, SO2) with a few exceptions in studies outside US that examined non-criteria pollutants (e.g., benzene, coarse particles, ultra-fine particles). Determining confounding by co-pollutants

64



1 2 3 4 5 6 7 8

that come from the same source(s) as NO2 is methodologically difficult, but it would be beneficial to measure, at least in a pilot study, the pollutants that may also have the health effects (including effect modification or synergism) or may be useful as markers to distinguish among the sources that emit NO2. When the EPA mentions “a multi-pollutant perspective”, I assume that there are several rationales for the idea. Ranking/prioritizing individual pollutants is difficult because there are uncertainties about the importance of specific pollutants, but they can be grouped in terms of the rationale for measuring them, and the rationale may be ranked.

Pollutant group Rationale (objective and feasibility) and comment

NO2, NO, NOx • Same instrument

CO • Marker of gasoline vehicles

• Multi-pollutant assessment using the instruments already used for NAAQS

• Potential health relevance

Black carbon • Diesel traffic marker

• Potential health effects

Ultra-fine particles / particle number concentrations

• Potential health relevance

• Expected to be high near-roadways

PM10-2.5, and their chemical constituents such as Br, Zn, Cu, Sb

Traffic: re-suspended dust, tire ware, brake ware.

PM2.5 and its chemical constituents such as EC, OC

• To determine “excess PM2.5” beyond urban background

• Potential health effects

SO2 • Sources other than traffic that produce NO2.

9

Charge Question 3 (Identifying Candidate Near-road Site Areas): 10

65



Comment: All of these items (a) through (f) (and population density, which was mentioned in the outline but not here) seem important to consider in identifying candidate near-road sites. However, without actually determining the relationships between these factors and the NO2 levels, presumably through the pilot project, it would be difficult to evaluate the adequacy of siting a monitor based on the information alone. EPA should look for studies that attempted to investigate these issues. The data from the New York City Community Air Survey (NYCCAS) may be useful. They have been conducting 2-week sampling measurements of NO2, SO2 (winter only), ozone, PM2.5 and its chemical constituents at 150 locations within New York City. They are conducting land-use regression of the measured pollutants including NO2 as a function of a number of geo-coded emission data including traffic volume and other local combustion sources (e.g., residual oil burning). The information from the analysis of the NYCCAS data may be useful in evaluating the limitation of AADT data.

1 2 3 4 5 6 7 8 9

10 11 12 13

Charge Question 4: Please comment on the available modeling tools (e.g., MOVES, AERMOD, etc.), and their pros and cons, that the subcommittee believes may be appropriate to discuss and/or recommend for use in the near-road monitoring guidance document.

14 15 16 17

Comment: I have not used these models and have not seen the model validation of these models as applied to NO2 and other traffic air pollution. Therefore, I cannot comment on this.

18 19 20

21

Charge Question 5: The use of saturation monitoring and on-road monitoring are also possible tools that state and local air agencies may choose to utilize in the near-road site selection process.

22 23 24

5 (a): What are the pros and cons to using passive devices to saturate an area to gather data?

25 26

Comment: The obvious pros include the low cost and small dimension. The obvious cons include the long sampling period required for the detection limit of the passive sampler. However, the spatial distribution of NO2 constructed from such sampling would be still useful in determining the siting of a sampler. The relationship between the 1-hr peak NO2 data and the data from passive samplers can be determined from a pilot study. The NYCCAS data mentioned above may be useful to do this, since the study already collected 2-years of data (to identify the high NO2 area) and the study is still going on (to measure hourly data at the high NO2 areas to compare the two-week vs. hourly data).

27 28 29 30 31 32 33 34

66



5 (b): What are the pros and cons to using non-passive devices, such as near real-time or continuous devices including, but not limited to portable, non-FEM chemiluminescence methods for NO2 or Gas Sensitive Semiconductors (GSSs) for NO2 and other pollutants of interest?

1 2 3 4

Comment: The pros: Ability to measure hourly data: The cons: Need to validate the correspondence with the FRM/FEM measurements. A pilot study is required for this.

5 6

5 (c): What would be the pros and cons, to a state or local agency attempting to use a specially outfitted vehicle to collect mobile measurements?

7 8

Comment: I am not sure what the pros would be unless all the other information leaves a kind of uncertainty that can be resolved by the mobile measurements. The cons include the cost and resources required to conduct the measurements.

9 10 11

12 13

Questions regarding the CO monitoring network and near-road monitoring Charge Question 6 (a): Does the subcommittee believe that the light duty cold start and congestion factors will significantly influence the location of peak CO concentrations in an area?

14 15 16

Charge Question 6 (b): Does the subcommittee have an opinion on whether, and possibly how, these two issues of vehicles operating under cold start conditions and light duty vehicle congestion and idling in urban street canyons and/or urban cores be considered in a future, nationally applicable, CO monitoring proposal?

17 18 19 20

Comment: I don’t know the data regarding the impact of cold start conditions on CO peaks and therefore cannot comment.

21 22

23 24

Questions regarding the PM monitoring network and near-road monitoring Charge Question 7: Does the committee believe that siting considerations for identifying the location of peak NO2 concentrations will likely address all of the high priority siting considerations for PM (particularly PM2.5) as well? If not, what other factors should be considered and what are the advantages in considering these factors for identifying the location of maximum PM concentration?

25 26 27 28 29

Comment: It depends on the region of US, but PM (particularly PM2.5) may be dominated by regional background PM levels, so the impact of the near-road pollution on the monitor will need to take into consideration (subtract) the data from non-NR PM monitor. I guess the NCore sites will be sufficient for this purpose where they exist.

30 31 32 33

67



Charge Question 8: In addition to PM2.5 mass, what other PM-related measurements are desirable at near-road monitoring stations (e.g., UFP number, black carbon, EC/OC, PM coarse, etc.)?

1 2 3

Comment: All of these would be “desirable”, but are funds available to measure these in the new near-road monitors?

4 5

6 7 8

Questions regarding the monitor siting criteria for microscale CO, microscale PM2.5, and the new near-road NO2 siting criteria Charge Question 9: Does the subcommittee believe that reconsideration of microscale CO siting criteria is appropriate? Specifically, would an adjustment of CO siting criteria to match those of microscale PM2.5 and microscale near-road NO2 sites be logical and appropriate?

9 10 11 12 13

Comment: I appreciate the historical background and the original reasons for the siting criteria for different NAAQS pollutants. None of these criteria of data collection were originally meant for epidemiological studies. However, since the researchers used these data from the regulatory monitors for observational epidemiological studies, and because the findings from these studies are in part influencing the process of setting NAAQS, it is inevitable that the siting criteria will need to accommodate the need to use the data for epidemiological investigation. These studies often use multi-pollutant regression models to examine potential confounding effects, which tacitly assumes that pollution variables equally represent the population exposures. The reported short-term associations between CO and mortality and cardiovascular morbidity raise a concern that these associations are observed despite the potential inadequacy of the exposure metric to represent population exposure (i.e., potential attenuation of associations). For this reason, I think it is appropriate to adjust CO siting criteria to match those of microscale PM2.5 and microscale near-road NO2 sites.

14 15 16 17 18 19 20 21 22 23 24 25 26 27 28

Charge Question 10: Even if the adjustment of microscale CO siting criteria in sections 2, 6.2, and table E-4 in

29 40 CFR Part 58 Appendix E to match that of microscale PM2.5

and microscale near-road NO2 is appropriate and proposed, should there be consideration to maintain the requirement on how urban street canyon or urban core microscale CO sites should be sited?

30 31 32 33

Comment: If we are assuming that there can be CO monitors for different purposes (and there is sufficient funds), yes, I think there should be consideration to maintain CO monitors that will measure the maximum impact that is separate from the population epidemiology.

34 35 36 37

68




Charge Question 11: Does the subcommittee have an opinion on how “urban street canyons” or “urban core” might be defined, perhaps quantitatively, and with regard to use in potential rule language?

1 2 3

Comment: Establishing such definitions would require some analysis of available data to characterize the relationship between the pollution levels and emission/environment conditions (AADT, building density, etc.).

4 5 6

7 8

Questions regarding the near-road monitoring pilot study

Charge Question 12: EPA and NACAA will select the locations for permanent sites … these considerations include choosing a large and a relatively small urban area based on population, an area with varied or complex terrain, an urban area with an operational NOX analyzer … Does the Subcommittee agree with these considerations? Further, are there other considerations that should be evaluated in selecting pilot cities to house permanent near-road monitoring stations as part of the pilot study?

9 10 11 12 13 14

Comment: Given the limited budget, EPA should consider the cities that already have infrastructure to conduct a pilot study or the cities that are already conducting multi-pollutant assessment at multiple locations. Atlanta and NYC come to my mind.

15 16 17 18

Charge Question 13: EPA and NACAA have proposed that at least two urban areas should have permanent near-road monitoring stations (that would fulfill NO2 near-road monitoring requirements) implemented for the pilot study… Specifically, what pollutants and other information should the pilot study measure or gather at the fixed, permanent monitoring stations, and by what methods?

19 20 21 22 23 24

Comment: See my comment on Charge Question 2 for the list of pollutants. In terms of other information any geo-coded information related to traffic and other emission sources would be useful. The reports from the NYCCAS project for such information may be useful (available from

25 26 27

http://www.nyc.gov/html/doh/html/eode/nyccas.shtml). 28 29

Charge Question 14: EPA and NACAA have proposed four to five urban areas to have saturation monitoring …Please provide comment on:

30 31 32 33 34 35 36 37 38 39

a. The pollutants that should be measured with the saturation devices at each saturation site. b. The number of saturation devices per pollutant, both passive and/or continuous/semi-continuous, that may be deployed in each pilot city.

69

http://www.nyc.gov/html/doh/html/eode/nyccas.shtml



1 2 3 4 5 6 7 8

c. Whether placing saturation monitoring devices near certain road segments should include, at a minimum: 1) the highest AADT segment in an area, 2) the road segment with the highest number of heavy-duty truck/bus counts, 3) at a road segment with more unique roadway design, congestion pattern, or terrain in the area, and 4) if feasible, at a lower AADT segment with a similar fleet mix, roadway design, congestion, terrain, and meteorology as the top AADT road segment in the area.

70



1 2 3 4

13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34

Mr. Rich Poirot Comments on Near Road Monitoring 1. The accompanying draft guidance document outline provides an initial 5 thought of the major topics required in the near-road monitoring guidance that will 6 aid state monitoring agencies in the identification and implementation of NO2 near 7 road monitoring sites from a multi-pollutant perspective. Please comment on the 8 overall content of the recommended topics in the draft outline. Please provide 9 suggestions on any missing subjects that should be included in the guidance 10 document and any unnecessary topics that are currently listed in the attached draft, 11 if applicable. 12 I think the draft outline for the guidance document seems reasonably complete and contains no unnecessary topics. One critical topic which seems missing from the outline is the importance of the specific distance from the roadway. I think “within 50 meters” is too broad a range, and that most of the near-road influence falls off within, rather than beyond, that distance. Extensive, long-term experience sampling near-road NO2 in the UK (where “roadside” monitors are within 5 meters from the road and where additional “kerbside” monitoring is conducted within 1 meter – both at heights between 2 and 3 meters) indicates that the “roadside increment” declines, predictably, with the log of distance from the road, as illustrated in the attached figure. From: NO2 Concentrations and Distance from Roads (2008) Air Quality Consultants Ltd. http://www.airquality.co.uk/laqm/documents/FallOffWithDistanceReptJuly08.pdf 35

36 37 38

This effect of distance is considered sufficiently predictable that a nomograph is available that estimates concentrations at any distance from measurements at any other distance (within 50 m). http://www.airquality.co.uk/laqm/tools/NO2withDistancefromRoadsCalculatorIssue2.xls While these estimates are derived from annual average concentrations, a similar relationship will occur for hourly near-road concentrations (of NO2 and other mobile source pollutants like BC, ultrafines, CO, etc.). Location changes within the 50 meter distance could easily result in changing the incremental roadway contributions by a factor of 2 or 3. I think there is a need to further constrain this distance range in the guidance, or

39 40 41 42 43 44

71

http://www.airquality.co.uk/laqm/documents/FallOffWithDistanceReptJuly08.pdf

http://www.airquality.co.uk/laqm/tools/NO2withDistancefromRoadsCalculatorIssue2.xls



1 2 3 4 5 6 7 8 9

10 11 12 13 14 15 16

17 18

perhaps the standard could be expressed in terms normalized to a specific distance (say 10 or 20 m). For ultrafine particles, especially those in the < 25 nm size range, the roadside gradient is likely to be even steeper. See for example: Zhu et al. 2004: “Aerosol Science & Technology: Seasonal Trends of Concentration and Size Distribution of Ultrafine Particles Near Major Highways in Los Angeles.” 38(suppl 1):5–13; Sioutas et al., 2005 Exposure Assessment for Atmospheric Ultrafine Particles (UFPs) and Implications in Epidemiologic Research. Environ Health Perspect 113(8): doi:10.1289/ehp.7939; Durant et al., (2010) Short-term variation in near-highway air pollutant gradients on a winter morning, Atmos. Chem. Phys. Discuss., 10, 5599–5626. I also think the emphasis on AADT as the primary focus for site selection is overstated relative to “other near-road considerations” which are likely more important (in addition to the specific distance) Note for example the following Table A3.6 from the 2007 Air Quality Expert Group report on Trends in Primary Nitrogen Dioxide in the UK, that maximum hourly (98th percentile) NO2 correlates poorly (R2 = 0.24) with counts of total vehicles, but shows a much higher correlation (R2 = 0.66) with counts of (diesel) bus traffic, based on measurements at 53 roadside sites in the UK.

From: UK air quality modelling for annual reporting 2007 on ambient air quality assessment under Council Directives 96/62/EC, 1999/30/EC and 2000/69/EC http://www.airquality.co.uk/reports/cat09/0905061048_dd12007mapsrep_v8.pdf 19

72

http://www.airquality.co.uk/reports/cat09/0905061048_dd12007mapsrep_v8.pdf



73

1 2 3 4 5 6 7 8 9

10 11 12 13 14 15 16 17 18 19

Long-term experience with roadside NO2 monitoring and modeling in the UK has also indicated that the enhanced dispersion effects of vehicle speed more than offset slight increases in NOx emission rates, leading to decreasing per-vehicle NO2 impacts with increasing AADT, especially on high speed “motorways” (analogous to US Interstates). Note the attached Figure 3.8 “adjustment factors” used to reduce per/vehicle NO2 emissions for modeling near-road NO2 impacts in the UK. From: Trends in Primary Nitrogen Dioxide in the UK (2007) Air Quality Expert Group Report, Annex http://www.defra.gov.uk/environment/quality/air/airquality/publications/primaryno2-20 trends/index.htm 21

http://www.defra.gov.uk/environment/quality/air/airquality/publications/primaryno2-trends/index.htm

http://www.defra.gov.uk/environment/quality/air/airquality/publications/primaryno2-trends/index.htm



1

3 4

6 7

9 10 11 12 13 14 15 16 17 18 19

29 30 31 32 33 34 35 36 37 38 39 40 41 42

Other minor suggestions include: • The relative age of the fleet may be an important component of the fleet mix, 2

which may vary among cities and within urban neighborhoods. (This may be somewhat less important for NO2 than for other near-road pollutants).

• The “load” on vehicles during rush hour(s) could be an important factor. For 5 example, diesel vehicles in stop-and-go traffic and/or on an uphill grade will lead to high NO2 regardless of AADT.

• The “expandability” of a site – i.e. the ability to accommodate additional samplers 8 for various other mobile source pollutants – should also be a site selection consideration.

• The availability (or establishment) of a “representative” urban background site (for NO2 and ideally for other MV pollutants) should also be an important consideration. Identification of “roadside increments” for the multiple pollutants is critical, as is the ability to project measurements from specific microscale sites to larger population exposures.

• The measured or expected neighborhood-scale background may also be important. Other things being equal, a high traffic road in the midst of other high traffic areas is likely to experience higher concentrations than a similar roadway on the edge of the urban area.

2. EPA and NACAA envision the near-road guidance document to be written 20 from a multi-pollutant perspective. What pollutants and sub-species does the 21 subcommittee believe should be included for consideration and discussion in the 22 near-road monitoring guidance? Some potential species for consideration include 23 NO2, NOX, NO, CO, PM (Ultrafine, 2.5, and 10), black carbon, air toxics (e.g., 24 benzene, toluene, xylene, formaldehyde, acrolein, or 1, 3, butadiene), and ammonia. 25 Please prioritize the recommended pollutants and provide the rationale for their 26 ranking, including how this pollutant measurement will contribute to scientific and 27 regulatory knowledge of near-road air quality and adverse human health effects. 28 Since NO2 is the focus of the revised NAAQS and new monitoring requirements, it (and the NO & NOx which typically come along with it) has to be given top priority. It seems possible however that of the many, various mobile source pollutants, NO2 may be one of the least health-relevant, and a new, large monitoring network measuring just NO2 would be a waste of scarce resources. All of the other pollutants listed above would also be of interest, but would be prohibitively expensive to add at all sites. I also wonder whether the near-road influence of these many mobile source pollutants (or even of just NO2 alone) is something that can be or should be addressed in a large network approach. Rather than prioritizing the above list and seeing a few species measured at a large number of (similar) sites, I would prefer to see a nested network, within which many/most of the above species could be added at a smaller subset of sites. Instrument costs and the availability of reliable, continuous samplers should also be an important consideration.

74



1 2 3 4 5 6 7 8 9

10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43

I think black carbon (preferably multi-wavelength) should be given high priority, given its relevance to health (& climate & visibility) effects, its strong influence – like NO2 – from diesel emission sources, and the availability of reliable continuous instruments. Continuous OC (OC/EC) data would also be useful at some sites to help assess effects of fleet mix, SOA formation, condensation, destruction, etc. If reliable instruments are available and affordable, particle number count (including ultrafine particle sizes) would also be an important measurement, and likely to increase rapidly near roadways (well within 50 meters). I would give BC and number count a higher priority than PM2.5 mass, and it seems likely that FRM/FEM PM2.5 samplers will substantially understate the semi-volatile fraction near roadways (although FDMS TEOMS have not generally performed well in the field). There is likely a steep PM coarse gradient near roadways, but PM10 (subtraction) measurements would be a poor way to characterize this. Collection of PM in different size fractions in large (aggregated) sample volumes – to support molecular level organic analysis and bioassay work - could also be useful at a few sites. CO measurements would be especially useful for contrasting pollutant mixes at sites (or times) with different diesel vs. spark engine fleet mixes. The “toxic” species listed above would likely show strong roadway increases, but are also likely to be prohibitively costly (or too labor intensive) to add at most sites. I wonder if there’s any possibility of moving or establishing one of the NAATS sites to a near-road location? Our (very small) VT state agency has recently had reasonable success operating a continuous BTEX instrument from Synspec – for which I believe a 1,3 butadiene option is available. Results from the Las Vegas MSAT near-road toxics study could be quite relevant here, and some consideration might be given to modifying planned future phases of that study to make it more relevant to the objectives of the new near-road NO2, CO NAAQS requirements and related multi-pollutant monitoring plans. I don’t know the availability, reliability or costs of continuous NH3 instruments, but better characterization of MV NH3 emissions would be desirable at a few sites at least. Possibly some periodic UC Davis DRUM sampling would be a useful complement at a few sites – if equipped with a streaker or somesuch to add time resolution to the ultrafines. I haven’t actually seen that configuration in action, and don’t know about current analytical capabilities at the DELTA Group. Ozone measurements might be useful at selected sites (including the urban background sites), as the contribution of secondary NO2 formation, even in near road environments, isn’t necessarily trivial, and interesting changes may occur with efforts to attain new ozone and NO2 standards. In addition to the above pollutant species list, and meteorological measurements, other measurements that should be considered include traffic counters (which can separate light & heavy duty MV) and or possibly cameras (which can be especially useful for evaluating extreme events). It should also be noted that the important objective of characterizing incremental roadway contributions for any of the above pollutants would benefit (as for NO2) from a measurements at a paired urban background site. Some (nearby) remote sensing (FEAT) could also be useful, or perhaps establishing near-road sites near locations where FEAT-type measurements have recently conducted (and may be periodically repeated).

75



1 http://www.feat.biochem.du.edu/assets/databases/Cal/Tricity_NH3_SO2_NO2_2008_Re2

5

7 8 9

10 11

13 14 15 16 17 18 19 20 21 22

24 25 26 27 28 29 30 31 32 33

35 36 37 38 39 40

41

43 44

port_ARB.pdf 3. Identifying Candidate Near-road Site Areas 3 a. AADT & Fleet Mix 4 As indicated above, I think AADT alone is a poor indicator. b. AADT & Fleet Mix 6 A metric which ‘diesel-weighted’ the AADT would be preferable to AADT alone, but again, count is not really the key issue, especially on high speed highways. Two trucks passing the monitor at 60 mph will not cause twice the impact of 1 truck at 30 mph… Also, given the 1-hour standard, the traffic and fleet mix on weekday morning rush hours are likely to be most important. c. Roadway Design 12 Assuming that you mean “no obstructions” between the road and the monitor, this seems reasonable, and it seems unlikely that that no suitable sites without such obstructions will be available. Barriers beyond the monitoring site that constrain the further dispersion of roadway pollutants should not be avoided and (in urban street canyons) may well lead to some of the highest population exposures. Conceivably, adding barriers - sound barriers, trees, etc. - might be considered as an exposure mitigation strategy. It might also be noted that large fractions of the population spend time within a 5 or so meters of congested urban streets, but population proximity to the edges of high-speed interstates with maximum AADTs is typically more distant.

d. Congestion Patterns 23 Conceptually, “level of service” sounds like an important indicator, although I don’t know how reliable such data is on a national scale. As indicated earlier, I think NO2 emission increases with speed are relatively small and offset by increased dispersion. I would expect higher concentrations during times/places of highest congestion, rather than during high speed driving conditions. e. Terrain Terrain could be an important, especially during the winter in mountain/valley locations, in urban street canyons, or near roadway dips which are below grade. e. Meteorology 34 I think the Agency’s proposed approach – strongly encouraging but not formally requiring “downwind” location is reasonable for all the reasons given. Ideally the “downwind” location would concurrently reflect the periods of highest traffic congestion and lowest wind speeds and mixing heights. In case of doubt, saturation sampling could help determine locations of maximum expected impact. Established sites which met measurements indicate are persistently upwind during rush hour should be replaced.

4. Modeling is another tool that may be useful in the identification of candidate 42 near-road sites. In particular, the use of mobile source emissions modeling with MOVES and local-scale dispersion modeling with AERMOD, can be presented

76

http://www.feat.biochem.du.edu/assets/databases/Cal/Tricity_NH3_SO2_NO2_2008_Report_ARB.pdf

http://www.feat.biochem.du.edu/assets/databases/Cal/Tricity_NH3_SO2_NO2_2008_Report_ARB.pdf



5 6 7 8 9

11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30

as part of the guidance document. Please comment on the available modeling 1 tools, and their pros and cons, that the subcommittee believes may be 2 appropriate to discuss and/or recommend for use in the near-road monitoring 3 guidance document. 4

Modeling may be a useful tool, but unless site-specific meteorology and vehicle mix, volume and congestion data are available, I’m not sure it would lead to a better site selection than a “common sense” approach. AERMOD also often performs poorly in complex terrain. 5. a. If a state were inclined to use saturation monitoring to aid in the selection of a 10

near-road monitoring site, and considering the NO2 standard is a 1-hour daily maximum standard, what are the pros and cons to using passive devices to saturate an area to gather data?

Unless the time periods for saturation sampling turn out to be atypical, the longer (than 1 hour) aggregation times for passive samplers may not be that big a problem. Based on the long-term, multi-site data from roadside sites in the UK, the peak hourly and annual average concentrations are well correlated across space, as indicated in Figure 3.3 pasted here. Note also the high correlation (R2 =0.98) between annual average NO2 98th percentile hourly values from 53 UK roadside sites in Table A3.6 above. From: http://www.airquality.co.uk/reports/31 cat09/0905061048_dd12007mapsre32 p_v8.pdf A disadvantage of passive samplers is that while reasonably accurate units are available for NO2, NO, NOx & BC, there may not be comparably accurate units for CO. This limits the ability to explore different NO2/CO ratios in a saturation approach. An advantage of passive samplers is their low cost and subsequent ability to deploy many units inexpensively. If passive sampling were limited to NO2 & NO, a relatively dense exploratory saturation sampling program could be considered, including innovative mounting of passive samplers on (and/or in) busses, commuter cars, taxis ect. This will open the can of worms regarding whether there’s intent to protect people in their cars at rush hour, but that issue probably needs to be addressed at some point. See also George Allen’s recommendations combining passive samplers with active, timed inlets.

33 34 35 36 37 38 39 40 41 42 43

77






5 6 7 8

10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44

b. Likewise, what are the pros and cons to using non-passive devices, such as near 1 real-time or continuous devices including, but not limited to portable, non-FEM 2 chemiluminescence methods for NO2 or Gas Sensitive Semiconductors (GSSs) 3 for NO2 and other pollutants of interest? 4

Potential advantages include the ability to collect data with higher time resolution and for more species than passive samplers allow. I don’t know the current instruments, and defer to others on the committee. c. Finally, what would be the pros and cons, to a state or local agency attempting to 9

use a specially outfitted vehicle to collect mobile measurements to assist in the near-road site selection process for NO2 specifically as well as other pollutants of interest?

Other than the prohibitive cost, this could be an excellent way to help select sites and characterize the space/time patterns of exposures to multiple roadway pollutants. I wouldn’t automatically rule this out, as it may turn out that some states or research groups may have access to such mobile sampling equipment and would be willing to operate it at reasonable costs. In my view the pilot studies should be conducted in fewer, rather than more locations, and this might be a good way to produce some useful data in a few (1 or 2) study areas. If only EPA had an adequately funded Office of Research and Development… There may also be some useful low-tech ways to combine “ordinary” mobile sampling vehicles (buses, taxis, commuter vans etc.) with passive samplers that could provide some useful information. 6. a. Does the subcommittee believe that the light duty cold start and congestion

factors will significantly influence the location of peak CO concentrations in an area? What priority should these factors be given when compared with the factors (AADT, Fleet Mix, Roadway Design, Congestion Patterns, Terrain, and Meteorology) already being considered for peak NO2?

Yes, these are important considerations. I don’t believe however that maximum NO2 and CO will necessarily occur at vastly different kinds of locations, and that some kinds of sites would be suitable for quantifying near-road influences from both pollutants. A relatively high fraction of diesel vehicles does not necessarily mean that emissions from spark engine vehicles will not be high as well. Congested sites where vehicle mixes change by time of day and day of week will be especially informative. Having concurrent, collocated data for multiple species, while searching for single pollutant “hot spots” is not likely to improve understanding of population exposures, help discern effects of co-varying pollutants, nor lead to development of effective abatement strategies. b. Does the subcommittee have an opinion on whether, and possibly how, these two

issues of vehicles operating under cold start conditions and light duty vehicle

78



1 2 3 4 5 6 7 8 9

10 11 12 13 14 15 16 17 18

719 20 21 22 23

24 W25 t26 t27 p28 r29 30 831

32 33

34 A35 m36 c37 l38 P39 i40 b41 c42 a43 c44

congestion and idling in urban street canyons and/or urban cores be considered in a future, nationally applicable, CO monitoring proposal? Are there other factors that may affect peak CO concentrations and not affect peak NO2 concentrations that should also be considered for any future CO monitoring proposal?

CO will also be influenced by residential wood combustion and other space heating emissions and so northern mountain valley locations with high traffic counts and congestion plus limited dispersion on cold winter mornings (when secondary NO2 formation is minimal) may see relaticely higher CO concentrations. As with NO2, I’m not convinced that CO is the most (or second most) health-relevant component of roadway emissions, and would hope that suitable near-road sites could be identified to address both pollutants, with a smaller number of sites added to address specific CO-specific concerns when the CO NAAQS revision is final. For both pollutants, I think the objective should be to characterize near-road population exposures to mix of traffic-related emissions, and not just to witch-hunt for the worst-case locations of maximum single-pollutant concentrations.

. Does the committee believe that siting considerations for identifying the location of peak NO2 concentrations will likely address all of the high priority siting considerations for PM (particularly PM2.5) as well? If not, what other factors should be considered and what are the advantages in considering these factors for identifying the location of maximum PM concentration?

hile there is likely a significant near-road enhancement of local PM2.5 concentrations, I hink this roadway enhancement is proportionally much smaller for PM2.5 mass - compared o the roadway enhancement of NO2, BC, ultrafines, etc., and that PM2.5 should not be a riority consideration in siting. Also, since diesel emissions are major contributors to oadway NO2 and PM2.5 there should not be much conflict in siting objectives.

. In addition to PM2.5 mass, what other PM-related measurements are desirable at near-road monitoring stations (e.g., UFP number, black carbon, EC/OC, PM coarse, etc.)?

s indicated above, I would give all of the above a higher priority than PM2.5 mass easurements, and would push more for continuous instruments that would better

haracterize the entire particle size distribution. Roadway fine particle concentrations are also ikely to include a substantial semi-volatile component, which is not well characterized by M2.5 FRM (or FEM) instruments. The roadway increment in coarse particle concentrations

s likely to be proportionately greater than for fine particles, and coarse-only sampling should e given a higher priority at some of these sites. Past consideration of setting an “urban” oarse particle NAAQS, was based on an assumption of greater inherent toxicity in urban reas. But this (logical) assumption was not supported by much measurement data. The arbon species (BC and/or EC/OC) and particle number information will be more useful than

79



P1 a2 k3 4 95

6 7

M2.5 mass for health effects studies and source attribution, especially given the longer veraging times – 24-hour and annual – for the PM2.5 NAAQS and currently stated intent to eep the annual standard “controlling”.

. To allow for near-road monitoring infrastructure to be multi-pollutant, and in

reflection of the recently promulgated near-road NO2 siting criteria, reconsideration of the existing microscale CO siting criteria presented in sections 2, 6.2, and table E-4 in 40 CFR Part 58 Appendix E may be warranted. Does the subcommittee believe that reconsideration of microscale CO siting criteria is appropriate? Specifically, would an adjustment of CO siting criteria to match those of microscale PM2.5 and microscale near-road NO2 sites be logical and appropriate?

8 9

10 11 12

13 A14 a15 v16 “17 518 n19 (20 b21 22 123

s previously indicated, I think the 50 meter range proposed for “near-road” NO2 is too large nd should be tightened prior to attempting to harmonize the various microscale criteria for arious pollutants. Conceptually, near-road measurements for multiple pollutants might standardized” to expected concentrations at a single fixed distance, or perhaps a (closer than 0 m) maximum and a minimum sampling distance could be specified, with a smaller umber of “research” sites encouraged that could collect useful data very close to roadsides inside the minimum distance) that would help characterize the roadway contribution without eing used for compliance determination.

0. Even if the adjustment of microscale CO siting criteria in sections 2, 6.2, and table E-4 in 40 CFR Part 58 Appendix E to match that of microscale PM2.5 and microscale near-road NO2 is appropriate and proposed, should there be consideration to maintain the requirement on how urban street canyon or urban core microscale CO sites should be sited?

24 25 26 27

28 O29 C30 31 132

33 34

35 N36

37 138

39 40 41 42 43 44

ff hand, I don’t see why near-road distances for CO and NO2 should be different, or why O should be relaxed to NO2 distances.


o opinion.

2. EPA and NACAA will select the locations for permanent sites that are part of the near-road pilot study based on which state or locals volunteer to participate and can process grant funds in a timely manner to deploy equipment. From this pool of volunteers, selection should be made on certain attributes that provide the best potential to fulfill pilot study objectives. In the attached draft white paper, EPA and NACAA have proposed some potential criteria for consideration in selecting where the fixed, permanent stations should be located. These considerations include

80



1 2 3 4 5 6 7 8 9 I10 w11 s12 a13 m14 i15 i16 t17 d18 t19 e20 e21 p22 t23 t24 25 126

27 28 29 30 31 32 33 34 35 36 37 38

39 T40 s41 N42 u43 a44

choosing a large and a relatively small urban area based on population, an area with varied or complex terrain, an urban area with an operational NOX analyzer representative of neighborhood or larger spatial scales for comparison to the near-road NOX analyzer, and an urban area with a cooperative (or non-cooperative) Department of Transportation. Does the Subcommittee agree with these considerations? Further, are there other considerations that should be evaluated in selecting pilot cities to house permanent near-road monitoring stations as part of the pilot study?

’m not sure sufficient funds are available to address so many different kinds of locations. It ill be necessary to take maximum advantage of existing sites (including urban background

ites and those operated by research groups) to the extent possible. The vailability/participation of academic or private sector groups to add supplemental easurements should also be encouraged. With such limited funds, is it necessary to

ntentionally include a small city or an area with a non-cooperative DOT? If small cities are ncluded (or not), selected sites should be adjacent to roadways with high volumes of diesel raffic and frequent rush hour congestion. Given the rapid decline of roadway influence with istance, priority should be given to sites which are substantially closer than 50 meters from he road (10 to 20 meters max). Possibly the effects of complex terrain could be more ffectively studied in a winter saturation study than with permanent monitors, although the xaggerated diurnal stagnation patterns and extent to which these correspond to rush hour eriods would be useful to characterize with continuous instruments. It would also be useful o consider urban areas which have different kinds of public transportation systems, bus and axi fleets and associated fuels, etc.

3. EPA and NACAA have proposed that at least two urban areas should have permanent near-road monitoring stations (that would fulfill NO2 near-road monitoring requirements) implemented for the pilot study. Please comment on the minimum equipment/pollutant measurement complement that should be deployed at each site and also the ideal equipment complement that each site should or could have, respectively. Specifically, what pollutants (e.g., NO2, NOX, NO, CO, PM (Ultrafine, 2.5, and 10), black carbon, air toxics (such as benzene, toluene, xylene, formaldehyde, acrolein, or 1,3, butadiene) and ammonia) and other information should the pilot study measure or gather at the fixed, permanent monitoring stations, and by what methods? This list should be in priority order, as feasible, and can include any NAAQS or non-NAAQS pollutant by any method (FRM/FEM and/or non-reference or equivalent methods), any particular type of other equipment for gathering supporting data such as meteorology or traffic counts.

he objectives seem to be somewhat mixed here. To a large extent the proposed pilot study eems to be focused on gaining insights into the process of citing near-road monitors (for O2 and to a lesser extent for CO NAAQS compliance determination. The emphasis is on nderstanding the relative importance of various traffic and roadway indicators (of varying nd often unknown quality) to guide NO2 site selection, the logistical and institutional

81



d1 p2 N3 t4 o5 e6 7 I8 r9 m10 h11 p12 g13 I14 15 116

17 18 19

21

23 24 25 26 27 28 29 30 31 32 33 34

36 37 38 39 40 41 42 43 44

ifficulties associated with establishing new sites in challenging environments, etc. From this erspective, the subsequent use of any resulting measurement data (other to confirm whether O2 and/or CO are exceeding or close to NAAQS) is almost irrelevant. Retaining several of

he fixed location sites, and building them into much more comprehensive sites where the bjective is to actually learn something about near-road multi-pollutant exposures is an ntirely different (but no less desirable) objective.

n selecting these few comprehensive sites, I would try to assure that they are close enough to oadways to capture the extreme gradients for pollutants like NO2 and ultrafines, and also ake sure there’s a relatively nearby representative urban site with similar measurements to

elp quantify the roadway increment. Because roadway emissions, and to a large extent opulation exposures in near-road locations tend to have large diurnal variability, I would enerally limit the measurements to species that can be quantified continuously. Beyond that defer to others on the committee to prioritize the species.

4. EPA and NACAA have proposed four to five urban areas to have saturation monitoring, using either passive devices and/or continuous/semi-continuous saturation type multi-pollutant monitoring packages (i.e., several types of monitors in one mountable or deployable “package”). Please provide comment on: a. The pollutants that should be measured with the saturation devices at each 20

saturation site. b. The number of saturation devices per pollutant, both passive and/or 22

continuous/semi-continuous, that may be deployed in each pilot city. I don’t have much expertise here, but think that (especially given the very limited budget), the selection of species and number of sites are inter-related and depend on available methods and costs. Ideally, the minimum species for saturation sampling would include at least NO2, NOx, CO, BC, but I don’t believe there are sufficiently reliable passive samplers for CO. Possibly a nested approach could be applied with larger numbers of passive NO2 samplers where applicable and smaller numbers of portable continuous devices for other key species. Note also George Allen’s suggestion to combine passive samplers with timed pump inlets – which might improve both pollutant sensitivity and temporal resolution. See also previous comments

c. Whether placing saturation monitoring devices near certain road segments 35

should include, at a minimum: 1) the highest AADT segment in an area, 2) the road segment with the highest number of heavy-duty truck/bus counts, 3) at a road segment with more unique roadway design, congestion pattern, or terrain in the area, and 4) if feasible, at a lower AADT segment with a similar fleet mix, roadway design, congestion, terrain, and meteorology as the top AADT road segment in the area.

All of the above seem like reasonable (but somewhat idealized) kinds of locations. I question whether it will really be possible to identify “a lower AADT segment with a

82



1 2 3 4 5 6 7 8 9

10 11 12 13 14

similar fleet mix, roadway design, congestion, terrain, and meteorology as the top AADT road segment in the area” or if in doing so it could realistically be assumed that differing AADTs were the sole cause of any differences in concentrations. The effect of differing AADTs might better be explored by sampling during different time periods along a single road segment. To the extent possible, it would be useful if these sites were located at similar distances from, and at rush hour downwind directions from the associated roadways. Assuming this may not be possible, lines of additional passive NO2 sensors might be added at each site, perpendicular to roadways and in upwind and downwind directions. Meteorological measurements may also be needed at some or all of these sites, and similar kinds of species measurements should be added a non-road representative urban site to help define the roadway increments from the different saturation sites.

83



1

5

9 10 11 12 13 14 15 16 17 18 19 20

22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43

Dr. Jay Turner 1) Please provide suggestions on any missing subjects that should be included in the 2

guidance document and any unnecessary topics that are currently listed in the 3 attached draft, if applicable. 4 The outline for the guidance document is fine.

2) Please prioritize the recommended pollutants and provide the rationale for their 6 ranking, including how this pollutant measurement will contribute to scientific and 7 regulatory knowledge of near-road air quality and adverse human health effects. 8 In general, the priority should be to monitor for indicator compounds for motor vehicle exhaust. In descending order of priority: (i) NO/NOx, opportunistic because the NO2 measurement method will likely by NO2 by difference (NOx minus NO); (ii) BC, as a second indicator for diesel emissions and thus sharpen the interpretation of the NO2 data; (iii) CO, a stronger indicator for gasoline-fueled vehicle emissions, to compare and contrast with the indicators for diesel emissions; (iv) air toxics, an indicator for vehicle emissions; and (v) PM as an indicator for diesel emissions, albeit with confounding by road dust (depending on the PM site that is monitored). For many of these pollutants there is the potential for confounding by high upwind concentrations – that is, the measurement (absolute concentration and/or concentration variations) may not necessarily be a dominated by vehicle emissions from the roadway.

3) Identifying Candidate Near-road Site Areas… 21 Is the objective to monitor at the highest NO2 site within 50m of a roadway, or at the site with highest NO2 attributed to the proximate roadway? That is, should there be consideration of aggregate upwind effects, e.g. from a dense roadway network, that leads to high background NO2 at the roadway to be monitored. Or, is the interest in selecting a roadway with high NO2 difference across the roadway? If it is the former case then an expanded list of considerations for identifying candidate sites is needed. In general, I support the development of a screening tool to guide the site identification process. Screening tools have been used for hot spot analyses and a similar approach could be used in the site selection process. One approach is to use dispersion modeling to create look-up tables to semi-quantitatively) relate roadway (and other sites) characteristics to potential impacts. A more refined analysis could subsequently be taken to prioritize the sites identified from the screening process. a) The proposed approach to consider AADT and fleet mix is reasonable as long as

there is also a consideration of vehicle speed (e.g. through the LOS, below) to capture the speed dependence of emission rates. Emission rate estimates based on more sophisticated and/or site-specific inputs could be used if the area has such information readily available, but is not a high priority.

b) The consideration of on-the-books vehicle emission controls is a low priority in the site selection process. Perhaps more important out-year considerations would be any programmed or planned changes to the roadway corridor.

c) A preference to sites at grade with no obstructions is a reasonable objective. There should be some flexibility, however, especially if the impact of other

84



1 2

4 5

7 8

10 11 12 13 14 15 16 17

19 20 21 22 23 24 25 26 27 28

30 31 32 33 34 35 36 37 38 39 40 41

43

designs or obstructions is such that a specific candidate site is still expected to a highNO2 concentration zone.

d) The consideration of LOS is best handled through a screening tool that is 3 grounded in dispersion modeling (see my preamble this question, above). The key is to capture the speed dependence of emissions.

e) I am less concerned for vegetation if the focus is on NO2 as long as the vegetation 6 does not lead to a significant airflow obstruction. This might be important factor, however, for other candidate pollutants.

f) Upwind pollutant meandering is certainly observed especially when winds are 9 light and variable (in which case, the meandering is driven wind direction variations) and/or the vehicle-induced turbulence dominates over the prevailing air flow. That said, preference should be given to site locations that are nominally downwind for the meteorological conditions leading to highest impacts. Perhaps a screening tool could be used to identify the conditions. A key aspect is overlaying the prevailing diurnal wind patterns with the diurnal traffic patterns to estimate conditions – and thus near-field locations – for maximum impacts.

4) Please comment on the available modeling tools, and their pros and cons, that the 18

subcommittee believes may be appropriate to discuss and/or recommend for use in the near-road monitoring guidance document. The use of MOVES and AERMOD might be overly burdensome for the site selection process. As previously described, I advocate a screening tool be developed that is grounded in dispersion modeling that provides semi-quantitative estimates of impacts. I agencies have MOVES outputs, which are link-based rather than trip-average and thus likely more relevant to the specific roadway environment, then this information could be used as input to the screening model. However, for many agencies it might be too burdensome to generate MOVES output for the near-roadway site selection process.

5) The use of saturation monitoring and on-road monitoring are also possible tools that 29 state and local air agencies may choose to utilize in the near-road site selection process… Saturation monitoring could be helpful in prioritizing candidate sites. There is substantial literature on saturation studies with passive monitors, but in these cases the integration times are typically long. The crux for this application is a saturation monitoring strategy that has sufficiently high time resolution to be relevant to 1-hour conditions (this does not mean that 1-hour resolution is needed). Given there is typically diurnal structure to both traffic patterns and dispersion conditions, one strategy is to collect samples for sub-daily time periods but integrated over several days (e.g. a battery of timer-based saturation samplers).

6) If EPA were to propose a new set of minimum monitoring requirements for CO near 42 roads, the near-road monitoring stations created under the implementation of the

85



3

7 8 9

10 11 12 13

15 16 17 18 19 20 21 22

24 25 26 27

29 30

32 33 34

36 37 38 39 40 41

43 44

I have no preliminary comments on this matter. 7) Does the committee believe that siting considerations for identifying the location of 4

peak NO2 concentrations will likely address all of the high priority siting 5 considerations for PM (particularly PM2.5) as well? 6 It might be adequate if the emphasis of the PM monitoring is on that component related to diesel exhaust emissions, and the emphasis in all of the monitoring is on roadway-specific impacts and not cumulative impacts which include consideration of upwind sources. With improved vehicle emissions control technology, the relative contribution of road dust, tire wear, and brake wear to the traffic-induced PM becomes more important. If the goal is to capture these impacts in the PM monitoring, this could lead to diverging siting considerations for NO2 and PM.

29. In addition to PM2.5 mass, what other PM-related measurements are desirable at 14 near-road monitoring stations (e.g., UFP number, black carbon, EC/OC, PM coarse, etc.)? While in the ideal case it might be desirable to monitor for various PM components, practical considerations likely make UFP number and black carbon the most reasonable candidates. UFP number can be highly variable and confounded by other atmospheric dynamics events. Thus, its measurement is most useful as site pairs across the roadway. Thus might be impractical. Black carbon would be of interest in its own right and to compare and contrast to NO2.

9) Does the subcommittee believe that reconsideration of microscale CO siting criteria 23 is appropriate? Specifically, would an adjustment of CO siting criteria to match those of microscale PM2.5 and microscale near-road NO2 sites be logical and appropriate? I have no preliminary comments on this matter.

10) Should there be consideration to maintain the requirement on how urban street 28 canyon or urban core microscale CO sites should be sited? I have no preliminary comments on this matter.

11) Does the subcommittee have an opinion on how “urban street canyons” or “urban 31 core” might be defined, perhaps quantitatively, and with regard to use in potential rule language? I have no preliminary comments on this matter.

12) Does the Subcommittee agree with the stated considerations for selecting sites for the 35 pilot study? The stated considerations are reasonable, although there relative weighting needs to be defined because some considerations are more important than others. Assuming resources will be limited, preference should be given to large urban areas with sufficient exiting monitoring infrastructure to place the road-side measurements in context.

13) Please comment on the minimum equipment/pollutant measurement complement that 42 should be deployed at each pilot study site and also the ideal equipment complement that each site should or could have, respectively.

86



1 2 3 4 5 6 7 8 9

10 11

13 14 15 16 17 18 19 20 21 22 23 24 25 26 27

The minimum additional measurements should include NO/NOx, CO, black carbon, meteorology (perhaps at a setback or other nearby representative location), and traffic characterization (vehicle count, class, and speed). CO2 could also be useful. It would be ideal to have each of these parameters (but certainly NO2) measured at a representative “background” site (background from the perspective of the roadway). The measurement matrix should take into consideration whether the data would be used to evaluate any screening tool(s) developed to aid in site selection. Beyond this minimum list of measurements, others could be added to fulfill specific study objectives. It is possible that research groups would be interested in adding measurement which would leveraging the investment in site infrastructure, and it would be great to accommodate this to the extent practicable.

14) Please comment on the saturation study design details. 12 The details are reasonable but it might be worthwhile to refine the saturation study objectives and then revisit the study design details. For example, is it desired to harmonize certain details within an urban area (or between areas, if possible) to more clearly evaluate the impact of other details? In one design the saturation monitors could be placed about the same distance from roadways, while in another design they could be placed at different distances from the roadways. If distance from roadway is a key parameter, then this could impact the data and its interpretation. Inconsistencies in upwind/downwind siting could strongly influence the interpretation of data from short-term saturation monitoring studies. If a screening tool was developed, one objective might be the evaluation of this tool through careful design of the saturation studies.

87



1 2 3 4 5 6 7 8 9

10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44

Dr. Warren H. White

AADT, fleet mix, roadway design, congestion, terrain, and meteorology all affect ambient concentrations, in complicated and interdependent ways. However, the combined near-road effect of all these influences can be described with just a few degrees of freedom. The framework outlined below is hardly new (e.g. White, 1977), but seems worth revisiting in light of the new rule.

The key to a simple description is that not much chemistry has a chance to occur

in the short time air spends near the road. A cross-road wind component of only 1 m/s, for example, carries air from 150 m on one side to 150 m on the other in just 5 minutes. On such a time scale the complex chemistry of smog formation can be considered determined by the surrounding air, independent of the fresh emissions. More precisely, the only reactions needing consideration are the rapid scavenging of O3 by NO [1] O3 + NO O2 + NO2, and the rapid photolysis of NO2 to yield [2] O2 + NO2 O3 + NO after additional steps. These reactions leave unchanged the concentrations of odd oxygen [Ox] = [O3] + [NO2] and nitrogen oxides [NOx] = [NO] + [NO2], and their relative rates establish a photostationary state that is generally fairly well approximated in the atmosphere:

[3] [O3][NO]/[NO2] ≈ k2/k1. Since Ox and NOx are chemically conserved near the road, their concentrations

respond only to physical dilution and mixing. They can be modeled as the sum of a variable contribution from roadway vehicle exhaust and a uniform background supplied by the surrounding air. For given concentrations [Ox]0 and [NOx]0 at the monitor, the reactive species can be expressed in terms of NO2: [NO] = [NOx]0 – [NO2] and [O3] = [Ox]0 – [NO2]. Substituted into the photostationary equilibrium [3], these identities yield a quadratic equation in [NO2] that can be solved for [NO2] in terms of [Ox]0, [NOx]0, and k2/k1. The following plots illustrate some features of the relationship.

The conservation of odd oxygen limits microscale NO2 maxima to the sum of

directly-emitted primary NO2 plus the reservoir of odd oxygen available in the surrounding air. An important siting consideration is therefore the middle-scale ozone background, which I did not see mentioned in the Study Approach or Charge Questions. This background bounds the NO2 produced from primary NO emissions, contrary to the impression one might get from statements such as this (FR v74, n134, 7/15/2009, p34441): “However, since the rate of conversion of mobile source NO to NO2 … is a generally rapid process, (i.e., on the order of a minute (ISA Section 2.2.2)), NO2 behaves like a primary pollutant in the near-road environment, exhibiting peak concentrations on or closely adjacent to roads.”

88



1 2 3

I will be happy to supply the spreadsheet used to generate Figures 1-3 to anyone else who might like to play with it.

4 5 6 7 8

9 10

Reference: White, Warren H. (1977) NOx-O3 photochemistry in power plant plumes: comparison of theory with observation. Environmental Science & Technology 11, 995-1000.

Figure 1. If oxidant background is 50 ppb (~ 50 ppb O3 PRB + <1 ppb NO2 PRB), then even 25% NO2 in the fleet exhaust and 200 ppb near-road NOx is not enough to make 100 ppb NO2.

11 12 13 14

89



1 Figure 2. The most favorable condition for NO2 at a typical background oxidant level (75 ppb O3 + 25 ppb NO2) is a dark sky (small photostationary ratio) to minimize NO2 photolysis.

2 3 4 5

90



1 Figure 3. In the absence of elevated exhaust NO2/NOx ratios, background oxidant is needed to convert the primary NO emissions.

2 3 4

91



1 Figure 4. Concentrations of conserved primary emissions like CO (“tracer”) fall off more rapidly with distance from the roadway than those of NO2 do.

2 3 4

92



1 2 3 4 5 6 7 8 9

10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40

Questions for EPA: I am not convinced that a substantial near-road monitoring program for NO2 and other traffic-related species is a good use of Agency resources. I think it will be hard to implement in a meaningful way, and I don’t see great potential value in the data it will produce. I recognize that the decision has been made already, and that I am not required to understand the reasons behind it. I could better focus on our charge questions, however, if I had answers for the following questions of my own. 1. What is meant by hourly NO2 concentrations – should they be equivalent to actual arithmetic averages of instantaneous concentrations? Exhaust concentrations at a near-road sampling inlet can vary greatly within a few seconds. In a given setting (background oxidant levels and meteorology), NO2 concentrations depend nonlinearly on exhaust concentrations. Under these conditions an instrument’s time response – and the nature of any ‘internal averaging’ – requires careful characterization. The reliance on a difference method (NOx-NO) further heightens the challenge for measurements near the road, where the signal/noise ratio is least. Is the goal, as it was with the PM2.5 FRM, to replicate the undefined and uncontrolled shortcomings of historical data that underlie existing epidemiological analyses? Or is it to make an accurate measurement of NO2? 2. How are concentrations from microscale locations to be linked to available public health statistics for epidemiologic analyses? Data from neighborhood- or urban-scale monitors have demonstrated utility for epidemiology because they are indicative of typical exposures for identifiable populations large enough to generate routine public health statistics. The numbers of residences near microscale monitors will be small, and the vehicle occupants driving by them will be anonymous. Will site-specific panel studies be required to connect the near-road data to health effects? 3. How large a slice of the monitoring pie is ultimately contemplated for near-road monitoring? The Agency deserves great credit for recognizing the need “to support measurement of multiple NAAQS pollutants” in calling this meeting. “Maximum expected hourly concentrations” are likely to occur at different locations (with different vehicle mixes and road characteristics) for different candidate species (e.g. NOx, CO, black carbon, PM0.1, and PM10). And health researchers will view the consequently different pollutant mixes as an important environmental signals for epidemiological analyses. Measuring different species at different sites would clearly be of little value for anything more than a NAAQS-compliance determination. Are we looking at NCORE on steroids, something like 75 x (number of traffic-related species) for the total number of sites?

93



1 2 3 4 5 6 7 8 9

10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44

Dr. Yousheng Zeng General Comments Near-road monitoring requirements: The purpose of near-road monitoring is to protect the health of residents living near roadways. There should be a screening criterion: In a particular CBSA, if there are no residents living within the 50-m corridor, near-road monitoring should be exempted. Following a similar line of thinking, if there is only one community within the 50-m corridor, the near-road monitor should be sited at this community, and not necessarily at a location where the impact is highest. In this case, other siting analysis is unnecessary. The end-point of near-road monitoring: Normally when an ambient monitor shows exceedance of NAAQS, state/local authorities are required to develop a State Implementation Plan (SIP) to bring the area into attainment with NAAQS. The State Implementation Plan will include some control measures to achieve attainment. If a near-road NO2 monitor shows exceedance of NAAQS, how will a non-attainment area be delineated and what does EPA expect the state/local authority to do? Due to the nature of significant concentration gradient along the roadways, the area with high NO2 concentrations could be extremely small. What will be the basis for designating an area as non-attainment area? The non-attainment is basically caused by mobile sources. In some areas, it is largely attributable to vehicles passing through the area on the interstate highways. What can the state/local authority do to achieve attainment? If the state/local authority cannot do anything, what is the point of requiring this type of near-road monitoring? EPA could conduct some studies and achieve attainment through regulations on vehicle emission standards. Charge Question 3.c In urban areas, the road segments that have high AADT are commonly elevated roadways. Requiring monitoring sites at-grade will either miss the plume from the roadways or significantly limit the choices for the monitoring sites. As far as the vertical location is concerned, the guidance document should consider the two factors – (1) the monitor’s probe intake should be in the general vertical area of the plume coming from the roadways; and (2) the residence time for sample to travel from the probe intake to the analyzer will meet the criteria (20 sec.), i.e., no extremely tall probe from the ground that cause a long residence time. As long as these two criteria are met, there is no need to specify whether the monitor needs to be at-grade. Charge Question 4 In many traffic related air quality impact analyses (e.g., air quality analyses as part of required NEPA process for highway projects), the CALINE3 and CAL3QHC models are used. They are still listed as preferred models on the EPA SCRAM webpage. EPA should evaluate these models along with AERMOD and provide guidelines on which model should be used for siting near-road monitors.

94



1 2 3 4

In this guidance document, EPA should explain if and how Ambient Ratio Method (ARM) and Ozone Limiting Method (OLM) can be used in conjunction with AERMOD model to convert freshly emitted NO to NO2. On June 28, 2010, EPA issued a memo addressing these issues for more general NO2 modeling (http://www.epa.gov/nsr/documents/20100629no2guidance.pdf). There should be some consistence between these modeling policy memos and this guidance document to be developed. If the guidance document identifies CALINE3 or CAL3QHC as allowed model, it also should explain if and how ARM and OLM can be used.

5 6 7 8 9

10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44

Also see my response to Charge Question 6 on modeling. Charge Question 5 A trailer-based transportable monitor will be very useful and practical for near-road monitoring. It will be self contained (a generator, analyzer, zero air, calibrator, retractable met tower, wireless modem, etc.) in a relatively small trailer. It can be pulled by a pick-up truck to a candidate site for a day, a week, or a longer period of monitoring. It will be moved to another candidate site. Once the candidate site screening is completed, the trailer can be stationed in the chosen permanent site, blocked up and tied down to serve as the permanent near-road monitor in that CBSA. The data generated by such a system will have the same quality as fixed monitoring station. Compared to a motor vehicle based monitor, the trailer-based unit offers comparable mobility at much lower cost, and it can used as a fixed monitor at a permanent site for years. In terms of data quality and comparability, the data generated by a FRM or FEM analyzer in the trailer-based monitor has a higher quality and confidence level than the data generated by other screening instruments (passive devices and portable instruments). Presumably there will be no meteorological (met) instruments collocated with passive or portable devices. An analysis of the data gathered by these devices will rely on met data from nearby met stations. For near-road monitoring, the wind conditions will be relevant and extremely localized. The analysis based on met data from some distance could be misleading. Charge Question 6 Peak CO concentrations are expected in urban street canyons and/or urban cores, especially at intersections where cars are idling in front of traffic light and the impact is coming from more than one street. I am not familiar with typical NO2 concentrations in this type of situation as compared to NO2 concentrations near major highway with heavy traffic. I am sure this type of data is available. If NO2 concentrations in urban street canyons are comparable to the NO2 concentrations near major highway, using one site to serve the monitoring need for both CO and NO2 should be encouraged in the guidance document. Otherwise, it would be infeasible to make a compromise between the two needs and the monitoring for CO and NO2 should be addressed separately. The CAL3QHC model is design to predict CO concentrations near road intersections. If EPA has validated the model, should the modeling be sufficient for determination of compliance with CO NAAQS and therefore no monitoring is required? In the recent SO2 NAAQS rule, EPA is changing its long-standing position of using monitoring data for NAAQS attainment determination, and will use modeling for NAAQS attainment

95

http://www.epa.gov/nsr/documents/20100629no2guidance.pdf



1 2 3 4 5 6 7 8 9

10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43

determination. For the same rationale, using a modeling analysis to determine NO2 and CO NAAQS attainment seems reasonable. Similar to (actually even worse than) the case of SO2, ambient CO and NO2 concentrations have an extremely high spatial variability near road. One monitor showing compliance with NAAQS at one street corner or road segment does not mean that the NAAQS is attained at different street corner or road segment. Modeling can cover a much larger space at a much lower cost. Even at the same street corner or road segment, moving the monitor by one meter could make the difference of attaining or not attaining NAAQS. Charge Question 11 Before responding to Charge Question 11, I would like to ask if monitoring these pollutants with extremely high spatial variability in a micro-scale is a good idea. See my response to Charge Question 6. If the answer is no, there is no need to spending resources to develop definition of “urban street canyons” and “urban core” and associated guidance for monitoring. In case EPA wants to pursue monitoring at street locations with high traffic volume and high spatial concentration variability, the following elements should be considered in defining urban street canyons:

• Traffic information similar to the one for near-road monitoring (e.g., AADT, posted speed limit, traffic light cycle)

• Street geometry o Ratio of street side building height to the width of the street (H/W ratio).

Need to develop an approach to the treatment of (1) different heights of buildings on the two sides of the street and (2) tiered buildings.

o One-way vs. two-way street (more plug flow in one-way and more turbulent in two-way street).

o Is the street lined with trees on the sidewalk? Tree canopy may have an effect of umbrella and trap portion of pollutants at the street level.

o Some way to normalize the H/W ratio with respect to number of traffic lanes on the street. This factor may not be important because the effect may have been incorporated by the combination of H/W ratio and the traffic volume (e.g., AADT).

• Meteorological factors: the angle between the street and prevailing wind direction (higher concentrations are expected if the angle is 90 degree).

In the context of ambient air quality monitoring rule, perhaps a set of cut-off values reflecting the above mentioned elements can be used to define urban street canyons. Charge Question 14.c. The impact of mobile sources to ambient air quality is governed by two types of factors, vehicle emissions and dispersion conditions. For compliance monitoring, the monitors should be placed near the highest impact area, which means both emissions and dispersion conditions are equally important. For the pilot study, however, factors associated with dispersion (e.g., terrain, roadway design, extremely micro-scale meteorological conditions) should be given more attention than factors related to the level

96



97

1 2 3 4 5 6 7 8 9

of vehicle emissions (e.g., AADT, fleet mix) because the emission rates can be characterized well using current tools (e.g., MOVES), AND the emission rates are the only parameter that impact ambient concentration in a linear or near linear fashion. If the pilot study can provide better understanding of the dispersion, the impact of a higher emission scenario can be anticipated or predicted by simply substituting the emission rates. The information derived from a more dispersion focused pilot study will be more useful than an emission focused pilot study.

teleconference on november 8, 2010. the appended...

Documents